ChangeLog

################################################################################
#
# Change log of IntaRNA available at 
#
# https://github.com/BackofenLab/IntaRNA
#
################################################################################

################################################################################
# changes in development version since last release 
################################################################################

# IntaRNA

## Interface and handling

- select and visualize the first rank-sorted CopomuS mutation automatically
  when browser results arrive (#317)

- select the first browser result and show its interaction visualization
  automatically; reuse CSV help text as result-header tooltips and store bundled
  examples in directly importable/exportable parameter JSON (#315)

- explain all browser input, CopomuS ranking and candidate-filter controls with
  tooltips; annotate compensatory mutations in one vaRRI view and compare all
  four-state energies in an adjacent Plotly chart (#313)

- isolate real-browser numerical fixtures from editable GUI examples and keep
  CSV export checks synchronized with selected columns and seed-end defaults

- integrate CopomuS as the browser Mutation Design task with single-pair input,
  native controls, extended tables and four-state vaRRI visualization (#307)

- fix IntaRNApvalue both-strand sampling and parameter consistency, bound retries,
  validate FASTA/fits and expose empirical p-values (PR #298)

- fix CopomuS explicit mutations, GC filters, signed indices, zero scores and
  shared profiles; validate inputs and preserve output on failure (PR #298)

- --outNumberPerSeed selects each seed's best candidate sites before final
  deduplication, energy/overlap/count filtering; model P ranks left-most-anchor
  ensembles; zero preserves ordinary prediction; the GUI exposes the quota next
  to the global output count (#285, #288 review)

- refine GUI seed and CSV control widths, filter examples by task, cap result
  table height and use dark active tabs for additional output views (PR #303 review)

- let users resize and transpose Plotly results, export their current dimensions,
  and highlight all reported seed regions in vaRRI (#305)

- compact and left-align GUI parameter rows, split CSV controls into their own
  tab, describe every additional output format and plot energy/probability files
  with sequence names, source tabs and PNG export (PR #287 review)

- reorganize GUI parameters, add conditional seed/accessibility controls and
  grouped CSV defaults, expand FASTA limits and refine result headers (PR #299)

- retain computed GUI runs as named, foldable sessions with JSON input restoration,
  independent outputs, per-session ZIP export and deletion (PR #297 review)

- permit clearing inherited CSV sorting when replaying JSON as text reports;
  validate JSON/GUI behavior in native release/debug and real browsers (#287)

- show prediction/evaluation tasks as always-visible choices with task-specific
  descriptions and an expanded Seeds/Accessibility heading (PR #287 review)

- upload or drop local FASTA files into each GUI strand with validation and
  protection against stale asynchronous reads (PR #287 review)

- clear all inputs on GUI reset and load named examples from maintainable JSON
  files through an Example menu (PR #287 review)

- load/store GUI settings as portable JSON; ZIPs contain parameter.json with
  embedded inputs/provenance and all outputs (PR #287 review)

- use one embedded-input JSON parameter document for each WASM invocation
  (PR #287 review)

- accept JSON parameter files with embedded text inputs and personality defaults
  through --parameterFile while retaining legacy configuration (PR #287 review)

- rename the browser interface directory and Pages URL to gui (PR #287 review)

- add selected-interaction vaRRI embeds and reproducible all-output ZIP downloads
  to the browser interface; compact parameters into accessible tabs and update
  the Freiburg group header and full-width result/footer notes (PR #287 review)

- extend the browser interface with vaRRI styling, prediction/evaluation tasks,
  multi-FASTA, start indices, CLI parameter groups, selectable CSV columns and
  additional output previews/downloads; label pair-relative differences mfe Δ
  and display actual compiled versions (PR #287 review)

- display sortable/filterable browser results and native CSV downloads;
  publish the tested interface alongside existing Pages documentation (#286)

- add responsive browser FASTA inputs, validation and abort dialog (#286, step 3)

- run browser predictions in cancellable, isolated Web Workers (#286, step 2)

- orient probability SVGs from the upper-left corner and highlight every MFE
  pair, splitting seed/MFE cell borders into orange and green (PR #259 review)

- enable approximate actual-pair CSV/SVG probabilities for unseeded and seeded
  heuristic ensembles, labelled explicitly as approximate (PR #259, step 11)

- reject nonfinite partition totals and finite-energy boundary underflow before
  publishing exact unseeded or heuristic pair probabilities (PR #259 review)

- enable base-pair CSV/SVG probabilities for exact unseeded model P/mode M
  predictions (PR #259, step 10)

- revise SVG axes, sequence labels and rounded tooltips; show intramolecular
  pairing (1-Pu) frames and the best interaction region (PR #259 review)

- BUGFIX : use each strand's accessibility energy scale for SVG frames and
  unpaired-probability exports, including mixed computed/imported inputs

- --out=bpsvg:FILE renders actual-pair probabilities with nucleotide/accessibility
  frames, a shared CSS palette, signed-coordinate guides, hover values and admitted
  seed outlines (PR #258, step 8)

- --out=bpProb:FILE writes actual-pair conditional probabilities for exact
  model P/mode M stack-seed ensembles, with disjoint-region merging, original
  coordinates and success-only publication (issue #257, implementation step 6)

- deterministic kinetic seed extension with --model=X --mode=K and configurable
  local scoring; complete-energy downhill steps, atomic noLP loop/stack moves
  and retained greedy traceback; equilibrium probability output is unsupported
- kinetic mode always uses stacked extensions and trusts handler-provided seeds;
  reuse candidate pairing/local energies
- IntaRNAsnap personality enables kinetic seed extension with noLP by default

- apply outDeltaE relative to the sequence pair's best interaction when merging
  regions; preserve local windows for outPerRegion output (PR #253)

- reject merged regions that cannot enforce the requested output overlap mode;
  retain independent selection with --outPerRegion (PR #253)

- clarify that suboptimal predictions have distinct interaction-site boundaries

- IntaRNAeval / --rri evaluates predefined RNA-RNA interactions (issue #184)

- compressed binary .agz accessibility caches for repeated screens (issue #245)

- too short sequences are skipped with warning if they are shorter than the 
  required seed or helix length (were causing an exception and abort before)
- default SHAPE method changed to "D", which is the default for ViennaRNA 2.7.*
- output buffer size increased to 512 KB for higher performance and reduced 
  file-handle overhead
- reject Zall/Eall-dependent output in window mode

## Technical changes and Optimizations

- run the existing CopomuS Python workflow in checksum-pinned, self-hosted
  Pyodide and replace subprocesses with validated JSON calls to IntaRNA WASM (#307)

- share Python companion-tool command execution, input validation and atomic
  output handling after rebasing onto IntaRNA-fuse (PR #298 review)

- reduce ordinary heuristic ensemble runtime by batching more source boundaries
  and avoiding exhaustive span probes for sparse seed domains (issue #301)

- record JSON/WASM/browser regression validation and preserve pinned JSON vendor
  checksums while marking upstream files as vendored for review (PR #287)

- build WASM with ViennaRNA 2.7.2 locally and the newest stable 2.x in CI;
  verify published archive checksums, isolate versioned caches and retain
  selected dependency provenance and notices (PR #292 review)

- BUGFIX : restore WASM dependency downloads using the pinned official ViennaRNA
  release asset; verify downloads before caching and recover invalid caches
  (PR #291)

- annotate inherited virtual methods/destructors with override, preserve compiled
  WASM asset provenance through staging, and expand protocol/browser validation
  for the PR #287 review

- cross-compile IntaRNA and pinned dependencies to WebAssembly (#286, step 1)
- verify browser predictions against native accessibility/energy references and
  exercise real WASM cancellation and nested Pages paths in Chromium/Firefox

- reuse MFE seed-extension scratch matrices and bound predecessor scans in exact
  and heuristic prediction; evaluate physical energies directly (PR #276 review)

- share exact left-most seed-anchor extensions with heuristic prediction,
  combining each true left extension with one retained right endpoint (#270)

- compute exact stack-seeded ensembles with left-most seed-anchored left/right
  extensions and shared bounded workspaces (PR #268 review)

- reuse maximal seed-extension workspaces for exact/heuristic partitions,
  endpoint selection and pair-probability differentiation (PR #266 review)

- compute exact and heuristic stack-seeded ensembles by first-seed ownership,
  seed-avoiding left prefixes and unrestricted right extensions; differentiate
  the same nonnegative retained ensemble for actual-pair probabilities (#264)

- initialize seed iteration outside the complete remaining sequence so a seed
  beyond a cropped region cannot hide an earlier valid seed

- validate probability-region dimensions, seed masks and overlap in release
  builds before accepting data or dereferencing a required annotation mask

- use hashed base-pair keys for sparse stack-seed lookup, assert internal seed
  geometry in debug builds, and document probability computation steps and
  contributor conventions (PR #258 review)

- document independent release/debug, installed-API and performance validation
  for the reviewed unseeded and heuristic probability outputs (PR #259)

- transfer completed probability-output buffers without copying; independently
  audit performance, accessibility conversion and recurrence depictions (PR #259)

- depict ensemble predictor recurrences and the exact base-pair inside/outside
  computation in internal SVG documentation (PR #258, step 9)

- validate exact pair probabilities on independent mathematical/native oracles,
  release/debug suites and installed consumers; document measured runtime and
  storage, and supply the required logging link dependency (issue #257, step 7)

- collect raw actual-pair masses and disjoint regional denominators with explicit
  successful-empty/nonempty/failed lifecycle (issue #257, step 5)

- count each exact stack-seeded interaction once; preserve tiny positive totals
  through complete updates and output forwarding (issue #257, step 4)

- select the checked suffix inside/outside kernel after independent correctness
  and maximal-stack timing comparisons (issue #257, implementation step 3)

- expose geometric stack-seed capability and compile contained occurrences plus
  deduplicated eligible-start boxes (issue #257, implementation step 2)

- establish independent weighted-chain fixtures for seeded pair probabilities
  (issue #257, implementation step 1)

- document the implementation plan for exact seeded base-pair probabilities,
  including algorithm comparison, numerical/output contracts and validation;
  no prediction or probability implementation changes (issue #257)
- require ViennaRNA >= 2.7.0 and its logger/prepared-window APIs at configure
  time; install the logger lock for every supported version (PR #267)

- fix ViennaRNA parameter-cache races and enable parallel prepared window
  accessibility; retain serialization for global MFE/PF and non-OpenMP
  ViennaRNA builds, with a source audit and concurrency regressions (issue #265)
- keep the concurrency regression compatible with Apple Clang's standard
  library using explicitly joined std::thread workers (PR #267)

- add an API overview, IntaRNA Doxygen styling, and automated documentation
  builds and GitHub Pages publishing alongside the user guide (issue #22)
  
- install personality links in out-of-tree builds, including IntaRNAsnap

- BUGFIX : normalize single-pair suboptimal boundaries before traceback and
  boundary-only output validation (PR #253)

- verify restricted heuristic traces against the independent tiny oracle and
  cover output overlap checks after outMinPu decomposition (PR #253)

- BUGFIX : apply terminal GU and accessibility output limits to later heuristic
  results, including seeded and helix predictors (PR #253)

- BUGFIX : heuristic ensemble suboptimals use finalized site energies and stop
  safely when no compatible candidate remains (PR #253)

- analyse output overlap modes and implement the reviewed regional guards,
  ensemble/output-filter fixes and global energy window; retain and document
  the one-right-extension limitation (issue #212, PR #253)

- serialize stored accessibility rows directly and load retained rows in place;
  benchmark direct/generic export and raw/gzip binary I/O (PR #250)
- add repository-specific AI coding guidance in AGENTS.md (issue #247)
- replace uBLAS storage matrices with std::vector and std::mdspan, retaining
  compact upper-band and triangular storage; configure selects native mdspan
  or bundled Kokkos headers while preserving GCC 14 support; document the
  public matrix operations and align their layout with repository guidance

- reduce temporary allocations and fix cleanup on error paths
- remove redundant accessibility checks from predictor, helix and seed handling
- C++23 required; build and installation checks with GCC 14 and Apple Clang
- avoid repeated partition-map lookups in exact ensemble prediction
- add exhaustive interaction and seed oracles for short-sequence tests
- reduced memory use for exact non-seed ensemble prediction without trackers
- BUGFIX : heuristic cell initialization and noLP ensemble recurrences
- BUGFIX : ensemble filtering and seed-extension partition reset
- BUGFIX : target accessibility limits and seed-free outMinPu filtering
- BUGFIX : accessibility range splitting and intramolecular ensemble energies
- BUGFIX : interaction assignment, seed comparison and output handling
- BUGFIX : dangling-end probability computation using maxLength+1 ED values
- BUGFIX : multi-threading : serialization of ViennaRNA-based accessibility 
  computation since vrna-lib not thread-safe (up to v2.7.2)
- BUGFIX : multi-threading : IntaRNAsTar was not thread-safe due to shared 
  storage of computed ranges

# IntaRNAsnap

New personality IntaRNAsnap identifies the fastest "folding path" to a (locally) 
optimal interaction after seed formation, rather than finding the global optimum 
under the assumption of thermodynamic equilibrium. 

To this end, IntaRNAsnap implements a kinetics-motivated greedy search for the 
best interaction that can be reached from the given seed(s) by a series of 
complete-energy downhill local moves. 
Local moves are direct stack extensions or 2-base-pair loop/stack moves, i.e. 
the same moves as used in the IntaRNA noLP model. That way, kinetic energy 
barriers posed by the formation of loops and bulges are avoided, which mimics 
the continuation of the zipping process after "jumping" over a loop or bulge.

This personality is only available with --model=X and --mode=K, and equilibrium 
probability output is unsupported. The local scoring can be configured with 
--kineticScore=A|B|C, and the output can include distinct structures, energies,
restricted partition sums, and trackers.


# IntaRNAeval

New personality of IntaRNA to evaluate predefined RNA-RNA interactions.
Given two RNA sequences and a set of predefined interactions, IntaRNAeval 
computes the energy and accessibility of each interaction, and reports the 
results in a CSV file. The evaluation can be performed with different energy 
and accessibility models, and the output can include distinct structures, 
energies, restricted partition sums, and trackers.


# Auxiliary Python output tools

- handle symlinked temporary directories in companion executable-lookup tests,
  including macOS /var aliases (PR #298)

- align companion tools on Python 3.10+, declare pip dependencies, install shared
  modules, and exercise source/installed/native commands in CI (PR #298)

- replace both R scripts with Python GEV/FDR estimation and region plotting;
  document optional dependencies, migration and regression tests (issue #295)

# IntaRNA_plotRegions.R

- fix deprecation warning

# CopomuS.py

- bugfix version checking

# IntaRNA-fuse.py

- replace the Perl prototype with an installed Python script, honor input
  sequences, validate external results, and document the fusion workflow (#294)

################################################################################  
################################################################################

261027 Alexander Mitrofanov
 * gui/copomus.js, scripts/wasm/test-browser.py
   * Select the first rank-sorted CopomuS mutation after populating the browser
     result table and immediately show its vaRRI and energy-plot visualizations;
     restore both after folding and reopening the session (issue #317).
 * gui/{index.html,docs/copomus.md}
   * Document the automatic initial mutation selection in the interface and
     CopomuS browser guide.

261027 Alexander Mitrofanov
 * gui/{results.js,sessions.js,index.html}, scripts/wasm/test-browser.py
   * Select the first energy-sorted interaction when a computation completes and
     load its vaRRI visualization; retain the selection through later table
     sorting and filtering.
   * Apply each CSV catalog description to both its result-header cell and sort
     button, matching the corresponding CSV-selection tooltip (issue #315).
 * gui/{app.js,examples/,tests/parameter-file.test.mjs}, gui/docs/
   * Move example display metadata into the example index so every example file
     is an ordinary Store/Load parameter document with no wrapper (issue #315).

261027 Alexander Mitrofanov
 * gui/{index.html,app.js,options.js,copomus-options.js,styles.css}
   + Add explanatory tooltips to every input label and all CopomuS ranking
     measures and candidate filters; set both FASTA input minima to 400 pixels.
 * gui/{copomus.js,visualization.js,index.html,styles.css}, gui/tests/,
   scripts/wasm/test-browser.py
   * Replace the four selectable mutation-state viewers with one wild-type vaRRI
     view that annotates both designed substitutions in dark green at their
     original signed coordinates.
   + Plot the ww, wm, mw and mm interaction energies beside vaRRI, including a
     dotted blue wild-type baseline and responsive stacked layout (issue #313).
 * README.md, gui/docs/copomus.md
   * Document the annotated single-view and four-state energy comparison.

261026 Alexander Mitrofanov
 * scripts/wasm/test-browser.py
   * Stop loading editable GUI examples as numerical browser-test fixtures; seed
     prediction, evaluation and mutation checks from explicit test-only inputs.
   * Derive the expected CSV download header from the rendered result columns and
     verify both seed boundaries, restoring the master WebAssembly workflow after
     the GUI default changed.

261026 Alexander Mitrofanov
 * gui/{index.html,app.js,styles.css,copomus*.js,runner.js,sessions.js},
   gui/examples/, gui/tests/, scripts/wasm/test-browser.py
   + Add Mutation Design after Prediction and Evaluation, enforce one unambiguous
     sequence per strand, and expose CopomuS measures, selection, filters,
     generators, thresholds and explicit encodings in the Sequences panel.
   + Retain ranked extended TSV output in sessions and compare wildtype, both
     single mutants and the compensatory mutant in vaRRI (issue #307).
 * python/copomus/{browser.py,workflow.py}, python/CopomuS.py,
   python/tests/test_copomus.py
   + Share CLI/browser ranking and output logic; run the existing mutation,
     filter and measure classes asynchronously against JSON prediction callbacks
     without changing CopomuS CLI behavior or output.
 * scripts/wasm/{build.sh,stage.sh,write-build-info.mjs},
   .github/workflows/build-wasm.yml, README.md, gui/docs/
   + Download and verify self-hosted Pyodide 314.0.7, stage the required Python
     sources and MPL notice, record runtime digests, and validate the complete
     Pyodide-to-IntaRNA-WASM workflow in Chromium and Firefox.
 * gui/output-visualization.js, gui/styles.css, gui/tests/output-visualization.test.mjs
   + Add resizable Plotly canvases and matrix-axis transposition controls.
   * Export PNGs at the current plotting dimensions instead of a fixed size.
 * gui/visualization.js, gui/columns.js, gui/tests/{export,protocol}.test.mjs
   + Pass every complete reported seed boundary to vaRRI as a translucent purple
     region highlight and include seed ends in prediction CSV defaults (#305).
 * gui/docs/step4-results-and-deployment.md
   * Document plot resizing, axis swapping, current-size export and seed overlays.

261024 Alexander Mitrofanov
 * python/tests/test_common_tools.py
   * Compare executable lookup against canonical fixture paths so macOS /var
     aliases do not cause false failures after the helper resolves symlinks.
   + Exercise sibling, source-tree fallback, missing and frozen executable
     lookup through an explicit directory symlink on every tested platform.
     Reproduce and verify the macOS CI failure using an aliased TMPDIR (PR #298).

261024 Alexander Mitrofanov
 * python/setup.py, python/pyproject.toml, python/MANIFEST.in, python/*/Makefile.am,
   python/Makefile.am, configure.ac, python/intarna_tools.py
   * Declare Python 3.10+ and NumPy/SciPy dependencies, offer optional plotting,
     and install shared helpers alongside standalone scripts and Python modules.
   * Replace legacy setup.py installation with an opt-in pip hook honoring DESTDIR
     without fetching dependencies or uninstalling packages in another prefix.
 * python/bin/build.py, python/intarnapvalue/test.py, python/README.md,
   python/bin/README.md, python/DEVELOPMENT.md, python/requirements-output-tools.txt
   * Make executable builds independent of the working directory and fail visibly;
     refresh documentation, source-test entry point and audit/performance notes.
   * Preserve direct invocation of the shuffle utility with argparse validation,
     optional sample counts and local random seeds.
 * python/copomus/measures.py, python/CopomuS.py, python/tests/*.py,
   .github/workflows/{build,python-output-tools}.yml, .gitignore
   * Validate shared profiles and thresholds; test malformed/NA profile data,
     installed lookup, atomic writes, native sampling and mutation workflows.
   * Exercise all companion tools and installation in Python/C++ CI, including
     the dependent PR base; exclude Python caches/build outputs (PR #298).

261024 Alexander Mitrofanov
 * python/intarnapvalue/intarna_pvalue.py, python/intarnapvalue/__main__.py
   * Independently shuffle the correct target and query in pairwise batches;
     apply the same parameter file to observed and shuffled predictions.
   * Bound favorable-interaction resampling with --max-attempts, fail immediately
     on command errors, and validate sample IDs/energies before publication.
   * Normalize wrapped single-record FASTA and DNA, validate counts, preserve
     zero energies, isolate RNG state and keep score ordering deterministic.
   * Evaluate fitted CDFs instead of numerical integration; retain historical
     raw-energy models, validate fits and expose empirical/none modes (PR #298).
 * python/intarnapvalue/dinucl_shuffle.py, python/tests/test_intarnapvalue.py
   * Handle singleton and invalid sequences; remove quadratic front-deletion
     from Euler-path construction and support instance-local RNGs.
   + Test composition, sampling, bounded errors, constraints and CDF equivalence.
 * python/intarnapvalue/README.md, python/Makefile.am
   * Document real options, conditional sampling, dependencies and API behavior.

261024 Alexander Mitrofanov
 * python/CopomuS.py, python/copomus/*.py, python/tests/test_copomus.py
   * Validate sequences/FASTA, mutation positions and bases, delimiters and counts;
     repair the explicit-selector call signature and CG/GC filter mismatch.
   * Preserve zero energies and quoted CSV identifiers; fix signed-index conversion,
     neighboring pairs across zero, and minimum-delta ordering at/below zero.
   * Share profile contents across accessibility measures, clean private temporary
     files, and reuse the wildtype prediction for all candidates (PR #298).
   * Report errors on stderr and replace output only after successful computation.
 * python/copomus/README.md, python/Makefile.am
   * Correct option names, defaults, Python requirements and mutation/ranking docs;
     distribute the focused regressions.

261024 Alexander Mitrofanov
 * python/intarna_tools.py, python/IntaRNA-fuse.py, python/IntaRNA_CSV_p-value.py,
   python/IntaRNA_plotRegions.py, python/Makefile.am, tests/testIntaRNAFuse.py,
   python/tests/test_output_tools.py
   + Share dependency-free executable lookup, checked subprocess calls, RNA/FASTA
     validation, CSV validation and atomic text publication between Python tools.
   * Rebase onto feat/294-python-intarna-fuse; retain both migrations, their
     installation rules and regression suites (PR #298).

261023 Alexander Mitrofanov
 * python/IntaRNA_CSV_p-value.py, python/IntaRNA_plotRegions.py
   + Replace the R tools with standalone Python scripts preserving positional
     arguments, negated-energy GEV tails, seven-digit rounding/BH correction,
     ordered region rows, inclusive coverage and plotting parameter files.
   + Validate input and fitting; publish CSV atomically, preserve original CSV
     fields, and support all previous figure formats without a display.
 * python/tests/test_output_tools.py, .github/workflows/python-output-tools.yml
   + Test analytical tails, known GEV samples, BH, CSV errors/round trips,
     plotting parameters, coverage and all seven output formats in Python CI.
 * python/README.md, python/requirements-output-tools.txt, python/examples/*,
   python/plotRegions.example.png, README.md, AGENTS.md
   + Document Python dependencies, migration, statistical and rendering
     differences, parameter defaults and examples; replace README R snippets.
 * Makefile.am, configure.ac, python/Makefile.am, R/*
   * Install/distribute the Python replacements, documentation and fixtures.
   - Remove the R scripts and their build directory.
   * Resolves https://github.com/BackofenLab/IntaRNA/issues/295

261023 Alexander Mitrofanov
 * python/IntaRNA-fuse.py, tests/testIntaRNAFuse.py, tests/runIntaRNAFuse.sh
   + Port the non-overlapping-site compatibility check and joint-accessibility
     energy calculation to Python's standard library, retaining the CSV layout,
     overlap criterion and both initiation terms.
   * Honor required target/query sequences instead of overwriting them with
     examples; handle empty predictions and reject failed commands or invalid
     output without silently replacing missing energies with zero.
   + Check sequence forwarding, constraints, fused energies, column selection,
     executable lookup and failures with subprocess fixtures and native tools.
 * perl/, Makefile.am, configure.ac, python/Makefile.am, tests/Makefile.am
   - Remove the Perl script/directory and their Autotools references.
   + Install/distribute the Python replacement and run its regression suite.
 * README.md, python/README.md, AGENTS.md
   + Document installation, usage, migration, output and energy semantics;
     update companion-tool paths for https://github.com/BackofenLab/IntaRNA/issues/294

261021 Alexander Mitrofanov
 * OutputConstraint, OutputHandlerSeedCandidates, PredictorMfe, seeded predictors
   + Add bounded per-seed candidate selection before global pruning, preserving
     eligible energy/traceback pairs and deduplicating sites before final filters.
   * Reuse each MFE mode with an admitted single-seed adapter; preserve ordinary
     statistics/trackers in one separate pass. Positive quotas can cost one
     prediction per seed; zero retains the existing path.
 * SeededPartitionFunction, PredictorMfeEns2d[Heuristic]SeedExtension
   + Rank/report left-most-anchor site ensemble energies independently of full
     site partitions, Zall and actual-pair probabilities, per #288 guidance.
 * CommandLineParsing, IntaRNA, README.md, doc/analysis/output-per-seed.md
   + --outNumberPerSeed=[0,1000], default zero; merge seed identities across
     regions/windows and reset per region combination with --outPerRegion.
   * Reject positive unseeded quotas, ignore evaluation, and apply final overlap
     constraints to the retained union without refilling quotas (#285).
 * gui/{options,protocol}.js, gui/tests/protocol.test.mjs,
   scripts/wasm/{smoke-protocol.mjs,test-browser.py}
   + Expose outNumberPerSeed in the first GUI output row between outNumber and
     outOverlap; move energy/accessibility filters to a dedicated second row.
   + Match native quota validation, window/region compatibility and JSON replay;
     cover the default, range, rendered order and real WASM execution.
 * PredictorSeedOracle_test.cpp, OutputHandlerSeedCandidates_test.cpp,
   runOutputPerSeed.sh, library/test Makefile.am
   + Independent chain-enumeration, traceback, seed-only, mode-matrix, merge,
     filter, CLI and unchanged-probability regressions; register/install sources.

261026 Alexander Mitrofanov
 * scripts/wasm/test-browser.py
   * decouple deterministic numerical browser checks from replaceable GUI example
     payloads while retaining real-browser coverage of prediction/evaluation loading

261026 Alexander Mitrofanov
 * gui/app.js, gui/{index.html,styles.css}, scripts/wasm/test-browser.py
   * place Explicit seeds beside a half-width Seed constraint selector and halve
     the CSV sort selector width; verify desktop/mobile control geometry
   * show only examples for the selected task, retain complete CSV downloads while
     scrolling result tables above 500 px, and match output tab selection styling
     to the parameter tabs (PR #303 maintainer review)
 * gui/docs/{json-workflow,step3-frontend-ui,step4-results-and-deployment}.md
   * document task-specific examples, compact controls and the bounded result table

261026 Alexander Mitrofanov
 * gui/{app,options,sessions,output-visualization}.js,
   gui/{index.html,styles.css}
   * compact bounded parameter rows while preserving their grouping and align
     wrapped labels above left-aligned controls
   * move CSV sorting/column selection into a dedicated tab between Output and
     Model & Energy; add content/use tooltips to all additional output formats
   + show sequence-aware Plotly profiles, heatmaps and selected-spot bars beside
     native file content; add header downloads and verified PNG export
 * .gitattributes, gui/vendor/, scripts/wasm/stage.sh,
   .github/workflows/build-wasm.yml
   + vendor and checksum Plotly.js Cartesian 4.1.2, stage its MIT license and
     verify both assets in the combined Pages artifact
 * gui/tests/output-visualization.test.mjs, scripts/wasm/test-browser.py,
   README.md, gui/docs/
   + cover signed categorical axes, file/sequence mapping, malformed data,
     responsive compact layout, plot rendering and PNG downloads in real
     Chromium/Firefox with the existing WASM engine
   * address https://github.com/BackofenLab/IntaRNA/pull/287#issuecomment-6416573560
 * src/IntaRNA/SeededPartitionFunction.cpp
   * Select sparse batch anchors by scanning admitted seeds instead of probing
     every possible pair in every source span; retain the grid lookup for dense
     domains and deterministic anchor order.
   * Let partition-only heuristic ensembles batch up to 16 MiB of hybrid
     coefficients, reducing repeated anchor-extension work without increasing
     exact or pair-probability table budgets (issue #301).
 * doc/analysis/{seed-extension-partition.md,heuristic-anchored-runtime.md}
   * Document the adaptive anchor lookup and targeted runtime/memory tradeoff.

261024 Alexander Mitrofanov
 * gui/{app,options,columns}.js, gui/{index.html,styles.css}
   * organize tabs/rows as Sequences, Seed, Output, Model & Energy, Query,
     Target and CLI-only; align wrapped labels and place Compute first
   + add a three-way seed selector, synchronized accessibility flags and
     mode-dependent input availability; place SHAPE beside accessibility
   * group compact CSV choices with descriptions and group selection actions;
     default to 10 predictions and expanded task-specific energy columns,
     explicitly preserving browser defaults in native-compatible JSON
 * gui/{protocol,uploads,parameter-file}.js, gui/examples/
   * accept 50 FASTA records per strand of 2-5,000 nt each, with longest
     selected query x longest selected target <= 1,000,000; allow 1 MB FASTA
   + warn about ignored FASTA header descriptions while using first-token IDs
 * gui/{sessions,results}.js
   * select results from the header/name, order Show/Hide before the name and
     ZIP/delete actions, compact table headings and distinguish selected rows
 * gui/tests/, scripts/wasm/test-browser.py, README.md, gui/docs/
   + cover limits, native JSON replay/defaults, dependent controls, grouped CSV
     choices, desktop/mobile alignment and result header selection
   * address https://github.com/BackofenLab/IntaRNA/pull/299#issuecomment-6375752155

261023 Alexander Mitrofanov
 * gui/{app,sessions,results,visualization}.js, gui/{index.html,styles.css}
   + retain successful runs in browser memory with unique IDs, editable names,
     independent results/diagnostics and JSON snapshots of their computed inputs
   * restore inputs and fold other output areas on selection; preserve sessions
     across edits, reset, imports, failures and aborts; unload folded viewers
   + export each session's original JSON/outputs from its header and delete it
     without overwriting editable inputs; keep controls keyboard accessible
 * scripts/wasm/{stage.sh,test-browser.py}, README.md, gui/docs/
   * stage the sessions module, exercise session lifetime and archive isolation
     in real browsers, and document temporary storage and updated reset behavior
   * follow-up to https://github.com/BackofenLab/IntaRNA/pull/297#issuecomment-6367326708

261023 Alexander Mitrofanov
 * .gitattributes, gui/docs/json-workflow.md
     + Preserve exact upstream license whitespace and mark the JSON header and
       license as vendored. Record native release/debug, WASM replay, browser,
       distribution and provenance validation for the PR #287 revisions.

261023 Alexander Mitrofanov
 * src/bin/CommandLineParsing.cpp, ParameterFile.h, tests/testParameterFile.py
     * Accept an explicitly empty CSV sort value to clear a parameter-file
       setting when reproducing non-CSV reports from the same JSON document.
 * gui/app.js, gui/styles.css, scripts/wasm/test-browser.py
     * Keep pending auxiliary-file reads from restoring reset inputs, return
       Reset to Sequence input and make the entire task label clickable.
 * doc/{Makefile.am,json-parameters.md}
     + Include JSON format documentation in source distributions and record the
       observed WASM footprint and reproducible dependency comparison (#287).

261023 Alexander Mitrofanov
 * gui/{index.html,app.js,styles.css}, scripts/wasm/test-browser.py
     * Replace the task dropdown with visible, highlighted native radio choices
       and task descriptions; update the heading to include Seeds/Accessibility.
     * Keep Example, Reset, Load and Store adjacent and test keyboard selection
       and responsive layout (PR #287 review).

261023 Alexander Mitrofanov
 * gui/{uploads.js,app.js,index.html,styles.css}, gui/docs/json-workflow.md
     + Add Upload controls and file-drop hints to both FASTA headers; validate
       file size/contents before replacing a strand and invalidate stale reads.
 * gui/tests/uploads.test.mjs, scripts/wasm/{test-browser.py,stage.sh}
     + Test local upload/drop, invalid files and reset during a pending read;
       stage the file-input module (PR #287 review).

261023 Alexander Mitrofanov
 * gui/{app.js,index.html,styles.css,examples/}, gui/docs/json-workflow.md
     * Reset clears inputs/results and restores defaults; example FASTA is shown
       only as placeholders. Add a described Example dropdown beside Reset.
     + Ship two editable JSON examples and a manifest, using the common importer.
 * gui/tests/parameter-file.test.mjs, scripts/wasm/{stage.sh,test-browser.py}
     + Validate/stage example files and check empty startup/reset behavior (#287).

261023 Alexander Mitrofanov
 * gui/parameter-file.js, archive.js, app.js, index.html, protocol.js
     + Load and Store non-default settings and embedded inputs using native JSON;
       validate imports before replacing the form and retain CSV selection order.
     * ZIP exports contain only parameter.json and outputs. Record version and
       native replay commands, including overrides for additional report formats.
 * gui/tests/, scripts/wasm/{smoke-protocol.mjs,test-browser.py,stage.sh},
   gui/docs/{json-workflow.md,step4-results-and-deployment.md}
     + Test JSON round trips, invalid imports, real WASM replay and output equality;
       stage the serializer and nlohmann/json license (PR #287 review).

261023 Alexander Mitrofanov
 * gui/protocol.js, gui/intarna-worker.js, gui/tests/protocol.test.mjs,
   gui/docs/step2-worker-architecture.md
     * Pass --parameterFile=/parameter.json to each WASM instance using the
       native JSON format with embedded FASTA, SHAPE and accessibility text.
       Preserve native output routing and independent report invocations (#287).

261023 Alexander Mitrofanov
 * src/bin/ParameterFile.{h,cpp}, CommandLineParsing.{h,cpp}, Makefile.am
     + Detect JSON parameter files; reuse native CLI validation and precedence,
       support embedded input text and retain files until computation finishes.
     * Honor parameter-file personalities when selecting native defaults.
 * src/bin/nlohmann/, doc/json-parameters.md, README.md, scripts/wasm/stage.sh
     + Vendor nlohmann/json 3.12.0 with its license, document the schema,
       dependency tradeoff and reproduction contract (PR #287 review).
 * tests/runParameterFile.sh, tests/testParameterFile.py, tests/Makefile.am
     + Check legacy/JSON equivalence, embedded inputs, precedence and errors.

261023 Alexander Mitrofanov
 * gui/, scripts/wasm/, .github/workflows/build-wasm.yml, README.md,
   doc/api-documentation.md, .gitignore
     * Rename webinterface to gui, including imports, staging, tests and the
       combined Pages artifact at /IntaRNA/gui/ (PR #287 review).

261022 Alexander Mitrofanov
 * scripts/wasm/{vienna-release.py,vienna-release.json}, build.sh
   + Resolve the newest stable ViennaRNA >=2.7.2 and <3 from all official GitHub
     release pages in CI; require its release tarball and published SHA-256.
   * Pin 2.7.2 for reproducible local builds and accept saved release manifests.
     Isolate RNAlib sources/installations and IntaRNA objects by archive identity
     and keep Boost/zlib in a separate prefix to prevent reuse or include-path
     shadowing of libraries/headers from another release (PR #292).
 * scripts/wasm/patch-vienna.py
   - Remove the dlib 19.24 compatibility patch; ViennaRNA 2.7.2 bundles dlib 20.0.
 * scripts/wasm/{write-build-info.mjs,stage.sh,smoke-protocol.mjs,test-browser.py}
   * Record the selected archive URL/checksum, check the compiled ViennaRNA
     version, stage matching dependency notices and test actual version exports.
 * scripts/wasm/test-vienna-release.py, .github/workflows/build-wasm.yml
   + Cover numeric version selection, pagination, major-3/prerelease exclusion,
     missing/invalid asset metadata, API failures, manifest validation and
     rejection of compiled engines with a stale ViennaRNA version.
 * webinterface/docs/step1-wasm-compilation.md
   * Document latest-2.x CI selection, pinned local reproduction, versioned
     caches, current bundled dlib and archive provenance.

261022 Alexander Mitrofanov
 * scripts/wasm/{build,download}.sh
   * Use the official GitHub ViennaRNA 2.7.0 release asset and pin its SHA-256;
     the university endpoint returned a short HTTP 200 text response, failing
     the WASM build before compilation (PR #291).
   + Verify archives before caching, discard invalid cached responses and remove
     failed partial downloads so rerunning a build can recover automatically.
 * scripts/wasm/test-downloads.py, .github/workflows/build-wasm.yml
   + Test verified-cache reuse, invalid-cache recovery, checksum failures,
     HTTP errors and truncated transfers against a local HTTP server in CI.
 * webinterface/docs/step1-wasm-compilation.md
   * Document the release source, archive verification and download regression
     command, including recovery behavior for failed or outdated caches.

261022 Alexander Mitrofanov
 * webinterface/{archive,visualization}.js
   + Build local UTF-8 ZIPs with normalized FASTA, compiled version provenance,
     portable parameter files, selected CSV, auxiliary inputs and all outputs;
     include separate replay configurations for native text report modes.
   + Construct selected-row vaRRI URLs from full target/query sequences, native
     dot-bracket output, names and signed indices, preserving strand direction.
 * webinterface/{app,results,protocol}.js, index.html, styles.css
   + Add row selection, embedded vaRRI/full-viewer link and Download ZIP action;
     preserve completed-run inputs and clear exports/viewer when inputs change.
   * Replace folded groups with keyboard-accessible tabs and compact parameter
     grids; link Bioinformatics - University of Freiburg to the group website
     and remove artificial width limits from table/footer notes.
 * scripts/wasm/{stage.sh,smoke-protocol.mjs,test-browser.py}
   * Stage the new modules; verify byte-identical replay through fresh engines
     and browser tab/selection/ZIP behavior in Chromium and Firefox.
 * webinterface/tests/export.test.mjs, README.md, webinterface/docs/*
   + Cover coordinate/identity mapping, effective settings and archive inputs;
     document replay commands, external visualization and browser limitations
     (https://github.com/BackofenLab/IntaRNA/pull/287#issuecomment-6321503136).

261021 Alexander Mitrofanov
 * src/IntaRNA/*.h
   * Mark 194 inherited virtual methods and 45 derived destructors override;
     keep algorithms, interfaces and third-party headers unchanged.
 * webinterface/{options,columns,protocol,app,results,intarna-worker}.js
   + Add CLI-based parameter groups and validation, multi-FASTA prediction,
     signed per-strand start indices and IntaRNAeval structure evaluation.
   + Provide all CSV columns, table visibility controls, pair-relative mfe Δ,
     local SHAPE/accessibility data and previews/downloads of native extra files.
 * webinterface/index.html, styles.css, assets/*
   * Follow vaRRI main-v2 colors and institutional header; keep one parameter
     group open at a time and adapt controls/defaults to the selected task.
 * scripts/wasm/{build,stage}.sh, write-build-info.mjs
   + Record actual compiled engine/library versions and asset hashes in text/JSON
     at build time; preserve/check provenance during staging and show it in UI.
   * Reject inconsistent missing overrides in the Emscripten build.
 * webinterface/tests/*, scripts/wasm/{smoke-protocol.mjs,test-browser.py}
   + Exercise task validation/defaults, real WASM evaluation/prediction, multiple
     records/coordinates, imported data, columns/reports and worker lifecycle.
 * .github/workflows/{build,build-wasm}.yml, README.md, webinterface/docs/*
   * Run WASM/browser checks on dependent PRs and document revised behavior,
     provenance, native option replacements and browser limits (PR #287 review).

261020 Alexander Mitrofanov
 * scripts/wasm/*, .github/workflows/build-wasm.yml, webinterface/app.js
   + Execute dependency API probes and test structured accessibility costs,
     native prediction references, C++ errors, nested paths and literal IDs.
   * Download archives atomically and clear obsolete diagnostics on input edits.
 * webinterface/docs/step4-results-and-deployment.md
   + Record native API/CLI, real WASM and Chromium/Firefox validation (#286).

261020 Alexander Mitrofanov
 * webinterface/results.js, app.js, index.html, styles.css, tests/*, README.md
   + Parse native CSV, preserve target/query orientation and inclusive indices,
     and show energy/relative suboptimality with sorting, search and export.
 * scripts/wasm/stage.sh, test-browser.py, .github/workflows/*
   + Stage matching WASM assets, notices and provenance; verify the real engine
     in Chromium and Firefox, including cancellation/restart and error recovery.
   * Publish one combined Pages artifact with the existing guide/API and the
     browser application at /webinterface/ (#286, step 4).

261020 Alexander Mitrofanov
 * webinterface/index.html, app.js, styles.css, mark.svg, docs/*
   + Add paired query/target editors, a valid example, bounded RNA/FASTA
     validation and result counts; normalize lowercase and DNA input.
   + Keep the form responsive during predictions, display errors/diagnostics,
     and abort by button or Escape (#286, step 3).

261020 Alexander Mitrofanov
 * webinterface/protocol.js, intarna-worker.js, runner.js, tests/*
   + Validate FASTA at the worker boundary, run the existing CLI with MEMFS
     input/output, and report CSV and bounded diagnostics asynchronously.
   + Terminate each worker on completion, failure or abort; test clean restart
     and stale-message handling (#286, step 2); document the worker protocol.

261020 Alexander Mitrofanov
 * scripts/wasm/*, .github/workflows/build-wasm.yml, configure.ac,
   src/bin/Makefile.am, src/IntaRNA/Makefile.am
   + Build pinned zlib, Boost and ViennaRNA for wasm32, then the existing
     IntaRNA CLI as an isolated ES module with MEMFS and callMain exports.
   * Omit pthread flags for single-thread Emscripten builds; retain native flags.
   * Resolve generated CLI headers/pkg-config data in the build tree; adapt
     bundled dlib Unicode and template code to the SDK libc++ and Clang.
   + Document the toolchain and validate native-reference predictions in WASM
     (https://github.com/BackofenLab/IntaRNA/issues/286, step 1).

261020 Alexander Mitrofanov
 * PredictorMfe2dSeedExtension.*, PredictorMfe2dHeuristicSeedExtension.cpp
   - Remove extensionLoopE(), the local loop-energy table, known-value flags,
     pair-index map and cache initialization/reset/fallback machinery requested
     in https://github.com/BackofenLab/IntaRNA/pull/276#discussion_r4312235144
   * Restore direct energy.getE_interLeft() and energy.areComplementary() calls
     in both predictors and traceback. Retain nonpreserving scratch resize,
     bounded predecessor scans and their existing traversal order.
 * PredictorSeedOracle_test.cpp
   - Remove obsolete cache probes; retain independent recurrence, workspace,
     tie, maximal-loop, range/offset, traceback and seed-witness regressions.
 * doc/analysis/mfe-seed-extension-reuse.md,
   doc/benchmarks/seed-extension/README.md
   * Describe direct evaluation and label the archived cached-prototype timing
     evidence explicitly so it is not attributed to the revised implementation.

261016 Alexander Mitrofanov
 * PredictorMfe2dSeedExtension, PredictorMfe2dHeuristicSeedExtension
   * reuse nonpreserving extension workspace capacity across seeds and traceback
   + cache seed-independent pair feasibility and loop energies within a bounded
     16 MiB budget per prediction; invalidate caches before changing offsets
   * bound predecessor scans without changing transition/update order (#274)
 * tests/PredictorSeedOracle_test.cpp
   + exercise repeated exact/heuristic predictions and cache correctness
 * doc/analysis/mfe-seed-extension-reuse.md, doc/benchmarks/seed-extension/
   + compare ensemble/MFE reuse strategies and record reproducible native
     runtime, memory and byte-exact output comparisons against 21b9d8f
 * doc/Makefile.am
   + distribute the analysis and benchmark evidence

261016 Alexander Mitrofanov
 * SeededPartitionFunction.*, StackSeedDomain.h
   * Share the exact left-most seed-anchor recurrence with heuristic mode.
     Remove the separate source-relative heuristic kernel; join true left
     extensions with the seed ending and one frozen selected right endpoint.
     Keep every right ending when there is no left extension. Crop right tables
     in batches without that no-left source and differentiate the same joins.
   * Document why pruning containing seeds preserves first-completed ownership,
     including shortest co-ending ties; retain singleton noLP support.
 * PredictorMfeEns2dHeuristicSeedExtension.*
   * Clarify shared factorization while preserving native right-selection hooks,
     energy ranking, complete-boundary publication and gapped-seed fallback.
 * SeededPartitionFunction_test.cpp, PredictorSeedOracle_test.cpp
   + Check nested ownership against the full declared family, one-sided paths,
     invalid/cropped selectors, singleton noLP support, batch recomputation and
     native virtual hooks against independent chain enumeration.
 * README.md, doc/analysis/seed-extension-partition.md,
   doc/analysis/heuristic-anchored-runtime.md, doc/benchmarks/seed-extension/*,
   doc/Makefile.am
   + Document the common recurrence and record reproducible kernel/native
     correctness, runtime, allocation and RSS comparisons. Add a native CLI
     runner with alternating samples and numerical pair-output verification.
   * Address https://github.com/BackofenLab/IntaRNA/pull/270#issuecomment-6216756696
     directly on perf/268-exact-seed-anchored-extension.

261016 Alexander Mitrofanov
 * StackSeedDomain.*
   + Enumerate a deterministic irredundant exact anchor family while retaining
     the original admitted occurrences for annotations and heuristic choices.
 * SeededPartitionFunction.*
   * Evaluate exact stack-seeded ensembles by left-most ownership in an
     equivalent seed family with redundant containing occurrences removed.
     Reuse anchor-local left/right and outside workspaces; combine complete
     boundary partitions in bounded batches before publication.
   * Differentiate the same nonnegative ensemble for actual-pair masses,
     including singleton support from either side under noLP. Preserve the
     heuristic's first-completed-seed ownership and gapped-seed fallback.
 * SeededPartitionFunction_test.cpp, PredictorSeedOracle_test.cpp
   + Compare nested mixed-length seeds, cropped/asymmetric spans and loop
     limits, noLP support and native boundary optima with independent chain
     enumeration. Check full hybrid callback coefficients, including zero
     boundary factors, and identical partition-only/pair publication.
 * doc/analysis/seed-extension-partition.md, doc/benchmarks/seed-extension/*,
   doc/Makefile.am
   + Explain exact seed ownership, extension bounds, outside differentiation
     and bounded boundary aggregation; record reproducible correctness,
     runtime, allocation and RSS comparisons against PR #268. Add correctness-
     only benchmark runs and record harness/shared-header source hashes.
   * Address https://github.com/BackofenLab/IntaRNA/pull/268#issuecomment-6212869312.

261015 Alexander Mitrofanov
 * SeededPartitionFunction.cpp
   * Allocate prefix, first-hit, right-extension and outside matrices once per
     computation at the inclusive span bounds. Use active upper index bounds
     and reset active cells before accumulation, including shrink/regrowth.
   * Reuse flat anchor/group storage and an endpoint index instead of creating
     per-group vectors and hash nodes; omit outside buffers in partition mode.
 * PredictorMfeEns2dHeuristicSeedExtension.cpp
   * Reuse the maximal right-extension buffer while freezing native heuristic
     endpoints, including explicit seeds of different lengths.
 * SeededPartitionFunction_test.cpp, PredictorHeuristicSeededProbability_test.cpp
   + Compare changing workspace shapes with independent chain enumeration and
     compare mixed-seed endpoint selection with isolated single-seed predictions.
 * doc/analysis/seed-extension-partition.md, doc/benchmarks/seed-extension/*,
   doc/Makefile.am
   + Document workspace bounds and reproduce exact/heuristic allocation counts,
     requested bytes, runtime, RSS and numerical comparisons against PR #266.
   * Address https://github.com/BackofenLab/IntaRNA/pull/266#issuecomment-6176325864.

261014 Alexander Mitrofanov
 * SeededPartitionFunction.*
   * Replace the combined absorbing-state recurrence by separate seed-free
     prefixes, canonical first-hit injections and shared right extensions.
     Apply full-boundary weights once after summing all seed owners; propagate
     pair masses through the same nonnegative recurrence. Co-ending variable
     seed lengths use their shortest admitted occurrence as canonical owner.
 * PredictorMfeEns2d{,Heuristic}SeedExtension.*
   * Route stack-seeded heuristic prediction through the common kernel with
     frozen right choices. Retain one-sided and selected-endpoint two-sided
     paths once; preserve legacy gapped-seed behavior and extension hooks.
 * SeededChainOracle.h, SeededPartitionFunction_test.cpp,
   PredictorSeedOracle_test.cpp, PredictorHeuristicSeededProbability_test.cpp,
   PredictorMfeEnsRegression_test.cpp
   + Check first-hit ownership, complete boundaries, raw pair masses and
     heuristic subsets with independent tiny-chain enumeration and fugacities.
 * StackSeedDomain.cpp, StackSeedDomain_test.cpp
   * Start iteration at the remaining sequence end; avoid colliding with an
     explicit seed outside a cropped active region. Cover offset and repeated
     regional predictions. Reject conflicting externally supplied lengths in
     release builds to keep seed-start and seed-end lookup consistent.
   * Align the conservative-handler capability test with the documented domain
     precondition in debug builds while checking retained seed geometry.
 * BasePairProbabilities.cpp
   * Restore public region/mask/overlap validation in release builds; the full
     suite exposed a missing-mask null dereference on the target branch.
 * README.md, doc/analysis/seed-extension-partition.md, doc/Makefile.am,
   doc/recursions/IntaRNA.SeededPartitionFunction.{inside,outside}.svg,
   doc/benchmarks/seed-extension/*
   + Explain the Step 6 theory transfer, numerical/structural constraints,
     changed heuristic semantics and validation; distribute the analysis and
     identify the superseded suffix-recurrence diagrams as historical.
   + Supply reproducible baseline comparisons of all boundary/pair masses,
     timing and process RSS, including the measured exact-kernel slowdown.
   * Implement https://github.com/BackofenLab/IntaRNA/issues/264 on its
     requested docs/257-base-pair-probability-plan base.

261013 Alexander Mitrofanov
 * StackSeedDomain, SeededPartitionFunction, PredictorMfeEns2dSeedExtension
   * use Interaction::BasePair coordinates and hashed sparse seed lookup tables
   * replace the stack-only handler's equal-length invariant with a debug assert;
     retain runtime validation of public input and numerical/extension failures
   + explain domain filtering, deterministic start traversal, inside/outside
     recurrence steps and coordinate conversion (PR #258 review)
 * BasePairProbabilities, AGENTS.md
   + explain validation and publication steps; document base-pair/hash storage,
     debug-only invariants and implementation-comment conventions
 * StackSeedDomain_test, seeded and heuristic probability oracle fixtures/tests
   + verify shortest shared seed endings, duplicate and unordered occurrences,
     row-major traversal and release-time rejection of invalid seed inputs
   * use the common base-pair representation while retaining independent oracles

261012 Alexander Mitrofanov
 * BasePairProbabilitySvg, BasePairProbabilityWriter, CommandLineParsing
   * restore top-to-bottom target indexing, aligned frames/guides and MFE region
   + reconstruct the MFE structure inside the best ensemble site for SVG only
   + highlight each MFE pair; shared seed pairs use orange vertical and green
     horizontal edges, preserving both annotations (PR #259 review)
 * testBasePairProbabilitySvg.py, README.md, seeded-probabilities-validation.md
   + verify complete internal MFE pairs, shared borders, signed guides and
     output-off parity; document the fixed-boundary traceback and orientation

261009 Alexander Mitrofanov
 * doc/analysis/seeded-probabilities-validation.md, doc/recursions :
   * record final review validation and controlled runtime/memory measurements;
     update the seeded-heuristic depiction for approximate pair output (PR #259)

261009 Alexander Mitrofanov
 * PredictorMfeEns2dHeuristic, PredictorMfeEns2dHeuristicSeedExtension,
   PredictorMfeEns2dSeedExtension, BasePairProbabilities, probability writers :
   + reverse retained candidate chains and signed seed-extension recurrences to
     obtain marginals of the existing heuristic ensemble; preserve denominator
     and seed-overlap corrections, validate all masses before publication
   + expose approximation metadata in CSV/SVG and retain seed provenance
   * bound cropped noLP continuations by the active region and correct default
     region offsets; reject unsupported singleton heuristic seeds explicitly
 * CommandLineParsing, tests, README.md, doc/recursions :
   + enable model P/mode H probability output; add independent tiny-chain and
     signed-extension checks plus CLI parity/metadata coverage (PR #259 step 11)
 * PredictorMfeEns2d, probability validation tests and report :
   * reject finite-energy boundary underflow and nonfinite heuristic denominators
     before publication; record independent review, full suites and performance

261009 Alexander Mitrofanov
 * PredictorMfeEns2d, CommandLineParsing :
   + compute actual-pair masses by reversing the existing unseeded recurrence;
     preserve site filters, noLP and numerical cutoffs, and merge disjoint regions
   * enable CSV/SVG with --noSeed; fix default-ended region offset truncation
 * tests/PredictorUnseededProbability_test.cpp, tests/Makefile.am, README.md :
   + compare against independent chain enumeration, filters, regions and failures;
     document the public optional sink (PR #259 step 10)

261009 Alexander Mitrofanov
 * BasePairProbabilitySvg, BasePairProbabilityWriter, CommandLineParsing, IntaRNA :
   * place target origin at bottom left and query name below the matrix; include
     nondefault sequence names in a subtitle and names in rounded hover values
   * show 1-Pu intramolecular pairing probabilities on the nucleotide frame
   + outline the best reported interaction region in green, including when
     ranked output is disabled; preserve ordinary output counts (PR #259 review)
 * README.md, tests/testBasePairProbabilitySvg.py :
   * document and cover coordinates, frame probabilities, tooltips and MFE bounds

261009 Alexander Mitrofanov
 * BasePairProbabilityWriter, BasePairProbabilitySvg, CommandLineParsing :
   * pass and validate separate target/query accessibility RT values for SVG
     frames; imported probabilities no longer inherit the interaction model RT
   * share the source-specific conversion with tPu/qPu, correcting their earlier
     physical-temperature conversion of computed base-pair accessibility (RT=1)
   * move completed SVG/CSV stream buffers instead of copying large documents
 * tests, README.md, doc/analysis/seeded-probabilities-validation.md :
   + cover imported Pu/ED, mixed sources, nondefault temperatures and invalid RT;
     compare against input probabilities and independent ED conversion
   + document independent review, controlled runtime/RSS comparisons and final
     release/debug, installation and public-header checks (PR #259)
 * doc/recursions/IntaRNA.PredictorMfeEns2dHeuristic.svg,
   doc/recursions/IntaRNA.PredictorMfeEnsSeedOnly.svg,
   doc/recursions/IntaRNA.SeededPartitionFunction.*.svg :
   * show the heuristic initial right-end GU gate and complete-site filters;
     place nonstack arrivals before inside aggregation and clarify outside notation

261009 Alexander Mitrofanov
 * doc/recursions/IntaRNA.PredictorMfeEns*.svg,
   doc/recursions/IntaRNA.SeededPartitionFunction.*.svg :
   + add eight standalone internal SVG depictions for unseeded exact/heuristic,
     seed-only, exact stack-seeded, legacy and heuristic seed-extension paths
   + show noLP transitions, seed admission and energy ownership, the production
     suffix inside/outside recurrences, actual-pair marking and regional merging
   + include implementation anchors and retain editable vector text; validate
     XML, glyph coverage and rendered layouts (PR #258 step 9, PR #259)

261009 Alexander Mitrofanov
 * BasePairProbabilityWriter, BasePairProbabilitySvg, CommandLineParsing :
   + add success-only SVG dot plots on files, gzip and synchronized standard
     streams; share CSV normalization and use original sequence coordinates
   + show sequence labels, 1-nt accessibility, seven CSS probability classes,
     legends, signed 10/50-nt guides, missing-zero boundary and hover values
 * BasePairProbabilities, PredictorMfeEns2dSeedExtension :
   + optionally retain admitted seed-cell masks with completed regional masses;
     annotate seed cells without changing probability or ranked-output semantics
 * tests, README.md, src/IntaRNA/Makefile.am, tests/Makefile.am :
   + document the SVG contract and register the writer and regression coverage
     for coordinates, seed admission, accessibility, empty/failed results and I/O
   + investigate the referenced dev-frank profile tracker files: their current
     versions contain profile output but no SVG implementation (PR #258 step 8)

261008 Alexander Mitrofanov
 * tests, README.md, doc/analysis, doc/seeded-probabilities-example.cpp :
   + finish probability regression and native performance gates, document the
     selected algorithm, supported ensemble, public API and numerical limits
 * CommandLineParsing, IntaRNA.cpp, OutputHandlerHub, SeededPartitionFunction :
   + retain copied output policy, check dense-grid dimension products, serialize
     seed-constraint initialization and emit empty matrices for too-short pairs
   + preserve dynamic OpenMP exceptions and report fatal failures as errors even
     with redirected warning logs; reject unsupported empty searches (PR #258)
 * IntaRNA.pc.in, .github/workflows/build.yml, Makefile.am :
   + link the logging dependency required by installed predictor consumers;
     compile/run a real probability consumer in CI and distribute new artifacts
   + GCC 14 release/debug: seven suites pass, 91,361 API assertions; all 74
     installed headers, installed consumer and source distribution checks pass

261008 Alexander Mitrofanov
 * BasePairProbabilityWriter, CommandLineParsing, IntaRNA.cpp, CLI/API tests :
   + add bpProb matrix output, capability/mode validation, multi-FASTA naming,
     gzip and synchronized standard-stream blocks; preserve spot coverage
   + guard OpenMP cancellation before publication, propagate write errors, keep
     tiny totals, support ranked count zero and mixed explicit seeds in debug
     validation (PR #258, step 6)

261008 Alexander Mitrofanov
 * BasePairProbabilities, PredictorMfeEns2dSeedExtension, API tests, Makefile.am :
   + add optional non-owning pair sink, original-coordinate raw regional commits
     and explicit success-only finalization; reject overlap and range failures
   + verify tracker parity, offsets, region crossings and injected failure after
     a successful region; pair-only computation retains no boundary map (PR #258, step 5)

261008 Alexander Mitrofanov
 * PredictorMfeEns*, OutputHandler*, IntaRNA.cpp, predictor oracle tests :
   + use the suffix kernel on the exact stack-only path, preserving virtual
     update hooks and the legacy heuristic/bulged paths; share the boundary
     objective with the outside pass and restore update scopes after exceptions
   + retain exact-zero status through collectors, hubs and ensemble diagnostics;
     test native Nussinov/nearest-neighbor boundary sums (PR #258, step 4)

261008 Alexander Mitrofanov
 * SeededPartitionFunction, PartitionArithmetic, StackSeedDomain, tests, doc/analysis :
   + add checked nonnegative kernels and raw pair differentiation; compare both
     against exhaustive and high-precision gold values and record time/storage
     tradeoffs before selecting the suffix backend (PR #258, step 3)

261008 Alexander Mitrofanov
 * SeedHandler*, StackSeedDomain, tests/StackSeedDomain_test.cpp, Makefile.am :
   + preserve each handler's admission rules, normalized allowances and retained
     explicit patterns; filter complete endpoints and span limits (PR #258, step 2)

261008 Alexander Mitrofanov
 * tests/SeededChainOracle.h, tests/SeededPartitionFunction_test.cpp, tests/Makefile.am :
   + independently enumerate weighted chains, raw boundaries and pair masses;
     verify overlapping witnesses and singleton/noLP semantics (PR #258, step 1)
   + preserve all ten high-precision theory fixtures with an explicit coordinate
     conversion and reproducible exporter; do not quantize their accessibility

261008 Alexander Mitrofanov
 * doc/analysis/base-pair-probabilities-plan.md :
   + specify the stack-seed partition and outside recurrences, sparse domain,
     suffix-automaton comparison, numerical policy, output integration and
     independent validation for https://github.com/BackofenLab/IntaRNA/issues/257
   + distinguish planned implementation from current supported behavior
 * README.md, doc/Makefile.am :
   + link the plan and include it in source distributions
261015 Alexander Mitrofanov
 * configure.ac, conda-build-env.yml :
   * require ViennaRNA >= 2.7.0, the first release with vrna_log_lock_set()
   * check header version and link the logger/prepared-window APIs during
     configure, including builds using --disable-pkg-config
 * IntaRNA/VrnaHandler, IntaRNA/AccessibilityVrna :
   * install the logger lock unconditionally for supported ViennaRNA releases
     and remove obsolete version guards; retain full-call serialization for
     RNAlib builds without OpenMP
 * README.md, AGENTS.md, doc/analysis/vienna-accessibility-thread-safety.md :
   * align the dependency and synchronization contract; distinguish historical
     2.4.14 source evidence from the supported range and recorded runtime tests
   + record minimum-version, release/debug and dependency-probe validation
   * addresses review of https://github.com/BackofenLab/IntaRNA/pull/267

261014 Alexander Mitrofanov
 * tests/AccessibilityVrna_test.cpp :
   * replace std::jthread, unavailable in the macOS CI standard library, with
     std::thread and exception-safe joins before inspecting worker results
   * fixes the Apple Clang regression build in PR #267
 * tests/runAccessibilityThreads.sh :
   * avoid empty-array expansion with nounset for older macOS Bash versions

261014 Alexander Mitrofanov
 * IntaRNA/VrnaHandler, IntaRNA/AccessibilityVrna,
   IntaRNA/InteractionEnergyVrna, bin/IntaRNA :
   * serialize shared parameter/compound setup, constraints and preparation
     through one exception-safe mutex, including native-thread library callers
   * run prepared window probability recurrences concurrently with OpenMP
     ViennaRNA; retain the conservative fallback for other builds
   * protect global MFE/PF compatibility state, remove worker-side global model
     resets, and install a separate logger lock for ViennaRNA >= 2.7
 * tests/AccessibilityVrna_test.cpp, tests/runAccessibilityThreads.sh,
   tests/Makefile.am :
   + compare serial/native-thread/OpenMP ED, ES and ensemble energies with
     varied models; compare CLI accessibilities/predictions with SHAPE D/W
 * doc/analysis/vienna-accessibility-thread-safety.md,
   doc/analysis/vrna-window-threads.cpp, doc/Makefile.am, README.md, AGENTS.md :
   + trace IntaRNA and ViennaRNA shared/owned state, provide an independent
     race reproducer and document supported concurrency boundaries
   * resolves https://github.com/BackofenLab/IntaRNA/issues/265

261007 Alexander Mitrofanov
 * doc/doxygen.cfg, doc/doxygen/* :
   + introduce the API main page with component links, prediction flow,
     coordinate/energy conventions, and library integration guidance
   + add IntaRNA header/footer navigation and responsive project styling
 * doc/build-api.sh, doc/check-api.py, doc/api-documentation.md, README.md :
   + build HTML without C++ dependencies and check entry-page links/assets
   + document local generation and the one-time Pages activation after merge
 * .github/workflows/documentation.yml :
   + build the existing Jekyll guide and API on PRs and master commits;
     publish both together from master only, with API content under /api/
 * Makefile.am, doc/Makefile.am :
   * rebuild documentation when headers or documentation assets change and
     include all documentation sources and build helpers in distributions
   * addresses https://github.com/BackofenLab/IntaRNA/issues/22

261005 Alexander Mitrofanov
 * bin/CommandLineParsing, tests/runKineticSeedExtension.sh :
   + add IntaRNAsnap personality with model=X, mode=K and outNoLP=true defaults;
     support executable-name and explicit personality selection
 * configure.ac :
   * find personality declarations via srcdir so out-of-tree installations
     create the executable links, including IntaRNAsnap
 * README.md, doc/kinetic-seed-extension.md,
   doc/recursions/IntaRNAsnap.PredictorSeedExtensionKinetic.svg, doc/Makefile.am :
   + document personality usage and depict seed initialization, allowed moves,
     complete-energy greedy selection, stopping and prefix reporting
   + distribute the benchmark's tutorial sequences with its script and results
 * doc/benchmark-kix.py, doc/kix-benchmark-20261005.json :
   + compare IntaRNAsnap with default IntaRNA with and without GU-end constraints;
     record wall time, peak RSS, MFE energy and maximum covered strand length
   * implement https://github.com/BackofenLab/IntaRNA/pull/254#issuecomment-5995041057

261005 Alexander Mitrofanov
 * IntaRNA/PredictorSeedExtensionKinetic :
   * always evaluate single and double stacks plus atomic loop/stack extensions;
     trust seed-handler structures/energies, including explicit lonely pairs
   * cache each end's candidates, shared complementarity checks and local
     energies; rebuild only the chosen end and refresh full opposite-end energy
 * bin/CommandLineParsing :
   * set --outNoLP=true with INFO for kinetic mode K
 * tests/PredictorSeedExtensionKinetic_test.cpp, tests/runKineticSeedExtension.sh :
   * update independent oracle and CLI expectations for atomic double stacks;
     cover trusted seeds, caching and kinetic mode K
 * README.md, doc/kinetic-seed-extension.md :
   * document revised semantics in response to
     https://github.com/BackofenLab/IntaRNA/pull/254

261005 Alexander Mitrofanov
 * IntaRNA/PredictorMfe, tests/PredictorMfeHeuristicCellState_test.cpp,
   tests/PredictorMfeEnsRegression_test.cpp :
   * normalize coincident suboptimal boundaries to one base pair before
     traceback or output, matching the initial optimum representation
   + assert valid single-pair candidates in traced and boundary-only output;
     fixes failures exposed by the PR #253 debug/sanitizer regressions

261005 Alexander Mitrofanov
 * tests/PredictorTinyOracle_test.cpp, tests/runOutputOverlap.sh :
   + verify every reported heuristic trace and energy against independently
     enumerated structures filtered by GU/ED limits; check forbidden intervals
   + exercise automatic outMinPu region splitting on both RNAs in all overlap
     modes with merged and independent output (PR #253)

261005 Alexander Mitrofanov
 * README.md :
   * retain and explain the one-right-extension strategy, including mode=M,
     incomplete compatible alternatives and the limits of blocking paired bases
 * doc/analysis/out-overlap.md, doc/analysis/out-overlap/reproduce.py :
   + record the accepted review decisions separately from historical findings
   * capture old output and new regional input rejections without treating
     historical bugs as regression expectations (PR #253)

261005 Alexander Mitrofanov
 * bin/IntaRNA, bin/CommandLineParsing, README.md :
   * filter merged results by the global minimum plus outDeltaE; keep the
     independent energy windows for outPerRegion=true (PR #253)
 * tests/runOutputOverlap.sh :
   + verify both region orders, zero/inclusive delta boundaries, permitted
     overlap modes, independent regional output and empty merged results

261005 Alexander Mitrofanov
 * IntaRNA/PredictorMfe, PredictorMfeEns, PredictorMfe2dHeuristic,
   PredictorMfe2dHeuristicSeed, PredictorMfe2dHelixBlockHeuristic,
   PredictorMfe2dHelixBlockHeuristicSeed :
   * share complete-site GU/ED validation between initial and later candidates
   * retain the existing matrix scan order and one-extension heuristic (PR #253)
 * tests/PredictorMfeHeuristicCellState_test.cpp :
   + cover all four matrix selectors, terminal GU and nonzero ED limits,
     all overlap modes, region offsets and boundary/traced output

261005 Alexander Mitrofanov
 * bin/CommandLineParsing, README.md :
   + validate manual regions and automatic decomposition against outOverlap
   + N requires one region per RNA, T one target region, Q one query region;
     B and independent outPerRegion output accept multiple regions (PR #253)
 * tests/runOutputOverlap.sh, tests/Makefile.am :
   + exercise all overlap modes, manual/automatic regions, per-region output,
     unchanged single regions and shifted sequence indices

261005 Alexander Mitrofanov
 * IntaRNA/PredictorMfeEns2dHeuristic :
   - remove raw-matrix suboptimal selection and its floating-point sentinel
   * reuse validated, finalized per-site energies and E_INF exhaustion from
     PredictorMfe; retain one best right extension per left boundary (PR #253)
 * tests/PredictorMfeEnsRegression_test.cpp :
   + compare N/T/Q energies against B with accessibility and energy cutoffs
   + cover exhausted and empty candidate sets with and without traceback

261005 Alexander Mitrofanov
 * README.md :
   + define suboptimal predictions by distinct start/end coordinates on both RNAs
   * distinguish site prediction from evaluation of supplied structures (PR #253)

261003 Alexander Mitrofanov
 * IntaRNA/PredictorSeedExtensionKinetic, src/IntaRNA/Makefile.am :
   + greedily extend seeds on either side using complete interaction-energy
     differences and deterministic thermodynamic or distance-weighted scores
   + support noLP macro-steps, GU restrictions, per-strand loop/span constraints,
     explicit-seed validation and cached trajectory traceback
   + retain valid visited prefixes and select non-overlapping output from the
     complete retained candidate set; reject unsupported ensemble statistics
 * bin/CommandLineParsing :
   + expose --mode=K exclusively for --model=X and --kineticScore=A|B|C
 * tests/PredictorSeedExtensionKinetic_test.cpp, tests/runKineticSeedExtension.sh,
   tests/Makefile.am :
   + validate local move choices, macro-step energetics, structural constraints,
     traceback, output filtering and CLI compatibility
 * README.md, doc/kinetic-seed-extension.md, doc/Makefile.am :
   + document scientific scope and complete-energy move enumeration

261002 Alexander Mitrofanov
 * doc/analysis/out-overlap.md, doc/analysis/out-overlap/reproduce.py :
   + analyse all four output overlap modes and the documented enumeration limits
   + reproduce regional overlap and energy-window violations, heuristic output
     filtering gaps, and invalid/incomplete ensemble candidate energies
   + provide 118 CLI observations, 64 controls and three independent site oracles
   + propose selection semantics and repairs for consultation; no predictor changes
   * addresses analysis request in https://github.com/BackofenLab/IntaRNA/issues/212
 * doc/Makefile.am :
   + distribute the analysis and standalone reproduction script

261001 Alexander Mitrofanov
 * IntaRNA/PredictorEvalOnly, src/IntaRNA/Makefile.am :
   + parse colon-separated hybridDB structures with sequence/index/pair validation
   + evaluate explicit base pairs with the selected energy and accessibility model
   + report distinct structures, energies, restricted partition sums and trackers
 * bin/CommandLineParsing, bin/IntaRNA :
   + --rri selects evaluation; IntaRNAeval personality requires interaction input
   + require one query/target and report ignored prediction and output constraints
   * bypass prediction result filtering to preserve distinct structures with equal
     energies, boundaries and base-pair counts; retain energy/accessibility settings
 * tests/PredictorEvalOnly_test.cpp, tests/runIntaRNAeval.sh, tests/Makefile.am :
   + check malformed structures, shifted indices, ignored filters, duplicate/tied
     structures, tracker coordinates, partition reset and energy contributions
   + reproduce ViennaRNA predictions through API and CLI evaluation round trips
 * README.md and CLI help :
   + document input format, retained settings and supplied-structure ensemble scope
   * resolves https://github.com/BackofenLab/IntaRNA/issues/184

260930 Alexander Mitrofanov
 * IntaRNA/Matrix.h :
   + const/mutable upper-band row views that exclude structural zeros and padding
 * IntaRNA/Accessibility, AccessibilityArchive, AccessibilityVrna, AccessibilityFromStream :
   * export stored ED rows directly without copies or per-cell getED() calls
   * retain generic export for other producers and constrained accessibility data
   * deserialize retained rows into final storage; validate discarded band tails
   * preserve version 1 archive bytes and existing constraint masking
 * tests/AccessibilityBinary_test.cpp, tests/Matrix_test.cpp :
   + verify row aliasing, byte-identical direct/generic output, zero interface
     lookups on direct export, constrained short sequences and corrupt discarded data
 * doc/benchmarks/accessibility.cpp, doc/benchmarks/accessibility.md, doc/Makefile.am :
   * compare direct/generic and raw/gzip binary I/O with exact ED verification
   + reproducible synthetic RNA and E. coli measurements with per-trial CSV results
   * retain gzip: both 100,000-base datasets shrink by 58-59%, despite slower I/O
 * README.md :
   + document direct matrix I/O and link the compression measurements
   * addresses review requests in https://github.com/BackofenLab/IntaRNA/pull/250

260929 Alexander Mitrofanov
 * IntaRNA/Accessibility, AccessibilityArchive, AccessibilityFromStream :
   + versioned Boost binary serialization of exact ED matrices and sequence data
   + stream rows with bounded scratch memory, retaining dangling-end intervals
   + validate sequence, dimensions, energies, archive format and stream integrity
 * bin/CommandLineParsing, IntaRNA/general :
   + select compressed binary accessibility input/output via .agz filenames
   * support all existing accessibility producers and both P/E input modes
   * release accessibility streams when parsing or writing fails
 * configure.ac, src/IntaRNA/Makefile.am :
   + check and link Boost.Serialization; distribute the internal matrix archive view
 * tests/AccessibilityBinary_test.cpp, tests/runAccessibilityBinary.sh, tests/Makefile.am :
   + exact round trips, malformed archives, prediction reuse and filename dispatch
 * doc/benchmarks/accessibility.cpp, doc/benchmarks/accessibility.md, doc/Makefile.am :
   + reproducible compressed text versus binary I/O benchmark with exact ED checks
 * README.md and CLI help :
   + document binary caches, unchanged text formats and compatibility requirements
   * resolves https://github.com/BackofenLab/IntaRNA/issues/245
 * configure.ac, IntaRNA/Matrix.h, IntaRNA/intarna_config.h.in:
   + prefer native mdspan, falling back to bundled Kokkos with a public
     INTARNA_USE_STD_MDSPAN define; allow explicit --with-mdspan selection
 * src/mdspan, src/experimental:
   + vendor only the Kokkos headers and both copies of its license from
     stable revision 8989f70749e28f337e6f7aa210db88659dba6f2f
   + distribute and install the headers with IntaRNA for external consumers
 * CI, README.md, doc/mdspan-storage.md:
   + restore GCC 14 release/debug builds and document automatic fallback
   + record a clean GCC 14/Kokkos comparison against uBLAS with all outputs
     matching across 160 executions; retain the original native measurements
 * IntaRNA/Matrix.h, AGENTS.md:
   + document public matrix operations and place explicit inline template
     definitions after class declarations, following the repository guidance
   + keep the guidance aligned with native/bundled mdspan selection (issue #246)

260928
 * IntaRNA/Matrix.h and storage aliases:
   + replace uBLAS dense, upper-band and upper-triangular storage with owned
     std::vector data accessed through C++23 std::mdspan
   + preserve logical resize, copy/move ownership and compact storage
 * configure.ac, README.md:
   + probe native mdspan library support and document the toolchain requirement
 * tests/Matrix_test.cpp, tests/benchmark/compare-matrix-storage.py:
   + add storage regression tests and reproducible before/after measurements
   + terminate the timed process group on timeout or interruption, with
     regression checks
   + include the biological benchmark inputs in source distributions
 * doc/mdspan-storage.md:
   + record validation and performance evaluation for issue #246

260928 Alexander Mitrofanov
 * AGENTS.md :
   + repository-wide AI coding guidance based on the current code and build setup
   + C++23, Doxygen and explicit inline definitions after class declarations
   + build/test commands, scientific invariants and two-part ChangeLog updates
   + separate implementation branches and PRs documenting targets, changes and results
   * resolves issue #247

260928 Martin Raden
 * IntaRNA/IndexRangeList:
   + listIsComplete : whether or not the list is complete (no further ranges can be added)
     + setter/getter
 * IntaRNA/Accessibility:
   * decomposeByMaxED() : fix issue #219
     + bugfix: requires serialization to be thread-safe due to shared storage of computed ranges
     + mark computed ranges as complete to avoid recomputation 

260925 Martin Raden
 * bin/CommandLineParsing :
   * isToShortQuery() : check if query length is < seedBP or helixMinBP (if required)
   * isToShortTarget() : check if target length is < seedBP or helixMinBP (if required)
   * parsing: 
     * provide warning instead of exception if query or target sequence is too short for seed or helix length
   * default:
     * default SHAPE methods changed to "D", which is the default for ViennaRNA 2.7.*
 * bin/IntaRNA :
   * skip too short query or target sequences 
 * IntaRNA/general :
   * newOutputStream() : 
     + increased output buffer size to 512 KB for higher performance and reduced file-handle overhead

260924 Martin Raden
 * IntaRNA/AccessibilityVrna :
   * fillByRNAplfold() : 
     + multi-thread bugfix: serialize access to ViennaRNA library since not 
       thread-safe (up to v2.7.2)

260921 Martin Raden
 * IntaRNA/general :
   * bugfix cleanup of stream objects
 * configure.ac :
 * src/IntaRNA/Makefile.am :
 * src/bin/Makefile.am :
   * undo some diagnostic flags to allow compilation on test system
 * IntaRNA/Accessibility* :
   * constructors: generate/read/keep data for maxLength+1 to allow for 
     dangling-end probability computation
 * tests/AccessibilityFromStream_test.cpp :
 * tests/AccessibilityVrna_test.cpp :
   + check for maxLength+1 ED values

260919 Alexander Mitrofanov
 * IntaRNA/PredictorMfe2d* :
 * IntaRNA/PredictorMfeEns2d* :
   * fillHybridE*(), fillHybridZ*(), predict(), parallelExtension()
 * IntaRNA/HelixHandlerNoBulgeMax : fillHelix(), fillHelixSeed()
 * IntaRNA/SeedHandler : isFeasibleSeedBasePair()
   * remove accessibility checks already covered by areComplementary()
 * tests/SeedHandlerNoBulge_test.cpp :
   + regression test for accessibility call counts

260919 Alexander Mitrofanov
 * configure.ac / m4 / .github/workflows/build.yml :
   * C++23 requirement, library checks and GCC 14 / Apple Clang builds
 * IntaRNA.pc.in :
   * fix installed include and library flags
 * IntaRNA/AccessibilityVrna / InteractionEnergyVrna :
   * C++ string constraints, avoid sequence copies and ensure cleanup on errors
 * bin/IntaRNA :
   * automatic cleanup of accessibility, energy, output and predictor objects
 * IntaRNA/OutputHandlerCsv / OutputHandlerEnsemble / OutputHandlerText :
   * use stream views to avoid temporary string copies
 * IntaRNA/PredictorMfeEns :
   * updateZ() : single lookup for partition insertion and accumulation
 * IntaRNA/RnaSequence :
   * use string::contains() for alphabet and ambiguity checks
 * tests :
   + exhaustive interaction and seed oracles for short-sequence tests
   + ViennaRNA constraint and repeated partition update tests
   * update Catch to 2.13.10

260919 Alexander Mitrofanov
 * IntaRNA/PredictorMfe2dHeuristic :
 * IntaRNA/PredictorMfe2dHeuristicSeed :
   * fillHybridE() : BUGFIX : reset best energy for each cell
 * IntaRNA/PredictorMfeEns2dHeuristic :
   * fillHybridZ() :
     * BUGFIX : reset best energy for each cell
     * BUGFIX : noLP double counting and missing stacked/bulged extensions
 * IntaRNA/PredictorMfeEns :
   * updateZ() : BUGFIX : apply noGUend and maxED filters
 * IntaRNA/PredictorMfeEns2dSeedExtension :
 * IntaRNA/PredictorMfeEns2dHeuristicSeedExtension :
   * predict() : BUGFIX : clear partition functions when no seed is found
 * IntaRNA/PredictorMfeEns2dSeedExtension :
   * fillHybridZ_left() : BUGFIX : multiply stacked partition factors
 * bin/CommandLineParsing :
   * parse() : reject Zall/Eall-dependent output in window mode
   * getTargetAccessibility() : BUGFIX : use target-specific length limits
   * getQueryRanges(), getTargetRanges() : BUGFIX : outMinPu without seeds
 * IntaRNA/Accessibility :
   * decomposeByMaxED() : BUGFIX : exclude forbidden positions and fix range length
 * IntaRNA/InteractionEnergyBasePair :
   * computeES() : BUGFIX : exclude empty structure
 * IntaRNA/NussinovHandler :
   * getQb() : BUGFIX : zero weight for out-of-range paired intervals
 * IntaRNA/InteractionEnergyVrna :
   * computeES() : BUGFIX : heed maximal base pair span
   * computeES(), computeIntraEall() : BUGFIX : free ViennaRNA data on exit
 * IntaRNA/Interaction :
   * operator=() : BUGFIX : handle self-assignment
   * operator==() : BUGFIX : handle missing seeds
   * Seed::operator<() : BUGFIX : compare all seed boundaries on energy ties
 * IntaRNA/OutputHandlerHub :
   * add() : BUGFIX : fix output forwarding signature
   * reported() : BUGFIX : report maximal child count
 * IntaRNA/OutputHandlerInteractionList :
   * add() : BUGFIX : handle zero storage capacity
 + tests : API and CLI regressions
 + doc/refactor/1-current-state.md : architecture and correctness audit

260919 Alexander Mitrofanov
 * IntaRNA/PredictorMfeEns2d :
   * fillHybridZ() : stream complete boundary partitions
 * IntaRNA/PredictorMfeEns :
   + updateCompleteZ() : preserve virtual updateZ() dispatch and restore mode on exit
   * updateZ() : consume complete partitions directly when no tracker is present
     (trackers, seeded and heuristic predictions retain buffered updates)
   + updateZisComplete : scoped mode flag; changed object layout requires C++ rebuild
   + addPartitionContribution() : shared filtering, conversion and Zall update
 * tests/PredictorMfeEnsRegression_test.cpp : streaming and buffered equivalence,
   trackers, overlap modes, virtual hook and exception recovery

250219 Martin Raden
 * python/CopomuS.py
 * python/copomus/IntaRNA.py
   * bugfix version checking (old string comparison fails for "3.12" < "3.7")

250214 Martin Raden
 + conda-build-env.yml : environment to build IntaRNA
 * new github build workflow

250213 Martin Raden
 * R/IntaRNA_plotRegions.R
   * fix deprecation warning when using element_lines() with ggplot >= 3.4.0

################################################################################
### version 3.4.1
################################################################################  

# IntaRNA
- BUGFIX : seed order check was buggy
- BUGFIX : explizit seed : wrong direction of query base pair parsing (thanks to Tim15-tech)

################################################################################  
################################################################################  

240918 Martin Raden
 * IntaRNA/SeedHandlerExplicit :
   * SeedData() :
     + BUGFIX : reversal of dotBar2 was missing

240430 Martin Raden
 * IntaRNA/PredictorMfeEns :
   * updateZ()
     * BUGFIX : no handling of partial Z including ED values (so far not used)

240429 Martin Raden
 * IntaRNA/Interaction
   * Seed::operator"<"
     * BUGFIX: less-than check fixed
   + operator<<(Boundary)
   + BasePair (explicit class definition)
 * IntaRNA/PredictorMfe :
   * updateOptima() :
     + no interaction update if "n==0"
 * IntaRNA/SeedHandler :
   * addSeeds()
     * BUGFIX : if internal seed base pairs were not found seed was still added

################################################################################
### version 3.4.0
################################################################################  

# IntaRNA
- new arguments t|qPfScale for sequence-specific ViennaRNA pf_scale parameter used
  to scale partition functions for ED value computation to avoid overflows
- BUGFIX : when using sequence subset selection in combination with precomputed
  accessibilities from file, wrong accessibilities were loaded for the selected
  sequences
- BUGFIX : seed-extension-based predictions (default prediction mode) were not 
  heeding accessibility constraints
- BUGFIX : query accessibility constraints were applied in reversed indexing

################################################################################  
################################################################################

240124 Martin Raden
 * IntaRNA/AccessibilityConstraint() :
   * AccessibilityConstraint() : 
     * BUGFIX: reversed index ranges were not stored
   * PredictorMfe2dHeuristicSeedExtension :
   * PredictorMfe2dSeedExtension :
   * PredictorMfe2dSeedExtensionRIblast :
   * PredictorMfeEns2sSeedExtension:
     * BUGFIX: missing isAccessible() checks added 

231201 Martin Raden
 * IntaRNA/RNASequence :
   + seqNumber : index of this sequence among its input sequence set. 
     relevant for accessibility loading from file
 * bin/CommandLineParsing : 
   * getFullFilename() : uses RNASequence::seqNumber instead of index within sequence
     vector. this enables the correct loading of respective accessibility files in 
     case only a subset of the sequences is used for prediction (--tSet or --qSet) 

230201 Martin Raden
 * IntaRNA/AccessibilityVrna :
   * constructor() :
   * fillByRNAplfold()
     + pfScale parameter
 * bin/CommandLineParsing :
   + t|qPfScale : sequence-specific parameters for explicit pf_scale setup for
     accessibility computation of long sequences
 * IntaRNA/VrnaHandler :
   * getModel() 
     + explicit pfScale parameter for rescaling of VRNA exp_params if pfScale >= 1.0
   + getPfScaleDefault() : provides default value from Vienna package
 * README.md :
   + docu of Q overflow and pfScale usage


################################################################################
### version 3.3.2
################################################################################  

# IntaRNA
- compilation fix: correcting shared variables for parallelization
- BUGFIX debug check
- BUGFIX IntaRNAhelix traceback of helices where seed is not at right end

################################################################################  
################################################################################

220913 Martin Raden
 * IntaRNA/HelixHandlerNoBulgeMax:
   * traceBackHelixSeed() : bugfix: tracing of helices with seed-trailing bps

220505 Martin Raden
 * bin/IntaRNA :
   * compilation fix: removing queryNumber from shared variables, since constant
 * IntaRNA/PredictorMfe2d:
   * predict() : bugfix: debug check was calling wrong function

################################################################################
### version 3.3.1
################################################################################  

# IntaRNA
- BUGFIX: empty lines with white spaces within FASTA input were causing parsing
  errors

################################################################################  
################################################################################

220328 Martin Raden
 * bin/CommandLineParsing :
   * parseSequencesFasta() :
     * BUGFIX: missing handling of empty lines with white spaces only within 
       FASTA input (were causing parsing errors) 

################################################################################
### version 3.3.0
################################################################################  

# IntaRNA
- `--outPairwise` enables pairwise sequence processing instead of all-vs-all
- input range for `--q|tIdxPos0` is now +-2,000,000,000 (was +-9,999,999) to
  enable genomic position settings

################################################################################  
################################################################################

220215 Martin Raden
 * bin/CommandLineParsing :
   + outPairwise : switch to trigger pairwise vs. all-vs-all sequence processing
   + getQueryNumberForTarget() : target-specific query sequence number
   + getQueryIndexForTarget() : index getter for target-specific query sequences
 * bin/IntaRNA :
   * query iteration now based on target-specific sequence number to enable
     pairwise sequence processing
220214 Martin Raden
 * bin/CommandLineParsing :
   * idxpos0 range now +-2,000,000,000 (was +-9,999,999) for genomic positions

################################################################################
### version 3.2.2
################################################################################  

# IntaRNA
- BUGFIX: maximal interaction length correction for precomputed accessibility
  data was one to large (wrong dangling end computation for maximally long RRIs)
  (thanks to Sabine Reisser)
- BUGFIX: explicit seed encodings within last 7 nucleotides (seedBP) were 
  ignored (thanks to Sebastian Holler)
- BUGFIX: outNoLP option was not correctly implemented (missing recursion cases)
  and was thus missing interactions
- BUGFIX: osx: configure adaptation to old grep version in osx

################################################################################  
################################################################################

220110 Martin Raden
 * configure.ac :
   * rewrite of grep personality call to be compatible with old grep version on osx

201204 Simon Bray (thanks!) :
 * replace travis with github action

210211 Martin Raden
 * IntaRNA/SeedConstraint :
   * constructor: bp>2 check only if no explicit seed present
 * IntaRNA/SeedHandlerExplicit :
   - getSeedMaxBP() : obsolete, replaced by getSeedMinBP()
   + getSeedMinBP() : minimal number of bps within encoded seeds; used for seed
     constraint initialization
   * traceBackSeed() :
     * bugfix: check for minimal seed length sufficient on one side
 * bin/CommandLineParsing :
   * seedBP now set to getSeedMinBP() if explicit seeds present
 * bin/IntaRNA :
   * exception information now motivates to send input along with report
 + docs/recursions : recursion depictions for sanity checks of implementations
   + PredictorMfe2d (+outNoLP)
   + PredictorMfe2dSeed (+outNoLP)
 * IntaRNA/PredictorMfe2dSeed :
 * IntaRNA/PredictorMfe2dHeuristic :
 * IntaRNA/PredictorMfe2dHeuristicSeed :
 * IntaRNA/PredictorMfeEns2dHeuristic :
   * bugfix outNoLP : missing recursion cases 
 * IntaRNA/AccessibilityFromStream : 
   * bugfix read from stream
   * test updated

201210 Martin Raden
 * IntaRNA/AccessibilityFromStream :
   * bugfix: max length == 1 smaller than max-accessibility-data-length due to 
     dangling-end treatment (thanks to Sabine Reisser)

################################################################################
### version 3.2.1
################################################################################  

# IntaRNA
- support for multi-seq input when `seedQ|TRange` given
- BUGFIX: Andronescu07 energy parameter set was not loaded by name without file
- BUGFIX: 'outMinPu' was not fully implemented
- BUILDFIX: non-global import of boost::bind

# CopomuS :
- exit value 0 if no favorable interaction was found

################################################################################  

201127 Martin Raden
 * bin/CommandLineParsing :
   * changing import and usage of boost::bind and boost::placeholders namespace
     (thanks to Behra Phani Rama Krishna)
 * R/IntaRNA_plotRegions.R :
   * replace deprecated expand_scale() with expansion() 

200615 Martin Raden
 * IntaRNA/OutputConstraint :
   + maxED : maximal ED penalty of each interacting subsequence
 * bin/CommandLineParsing :
   * getOutputConstraint() : 
     + energy argument to setup OutputConstraint constructor call
 * IntaRNA/Predictor :
   * updateZall() :
     + check for OutputConstraint.maxED
 * IntaRNA/PredictorMfe :
   * updateOptima() :
     + check for OutputConstraint.maxED
 * IntaRNA/PredictorMfe* :
   * obsolete OutputConstraint variable removed

200602 Martin Raden
 * bin/CommandLineParsing :
   + sequence error information now with sequence ID rather than number

200602 Martin Raden
 * bin/CommandLineParsing :
   + explicit seedQ|TRange check for each sequence (support for multi-seq input)

200429 Martin Raden
 * python/copomus/candidate_selectors.py :
   * bugfix: sys.exit(0) if not favorable interaction was found

200312 Martin Raden
 * bin/CommandLineParsing :
   * validate_energyFile() : 
     + bugfix: missing check for predefined Andronescu07 data set name

################################################################################
### version 3.2.0
################################################################################  

# IntaRNA
- BUGFIX: accessibility blocking constraints were only applied to seed location
- improved Zall estimate (and depending values) for `--model=X` (default)
- new arguments:
  - `qId|tId` : optional id (FASTA prefix) setup for sequence naming
  - `acc|accW|accL` : meta accessibility setup for both query and target
  - `intLenMax|intLoopMax` : meta interaction and interior loop length setup
  
# CopomuS
- Compensatory mutation selector to support interaction validation experiments

################################################################################

200302 Martin Raden
 * IntaRNA/IndexRange :
   * fromString() :
     * bugfix: + missing parsing of negative indices 
 * IntaRNA/InteractionEnergy :
   * areComplementary() : 
     + check if both positions are accessible (to avoid additional checks in
       predictor recursions)
 * bin/IntaRNA :
   * ambiguous nt warning now in verbose log (was info log) 

200217 Martin Raden
 * bin/CommandLineParsing :
   * NumericParameter : 
   * CharParameter :
     + isSet() : checks if value and default differ
   + qId|tId : id (prefix) for sequence naming
   + acc|accW|accL : meta for q|t*
   + intLenMax|intLoopMax : meta for q|t*
   * resetParamDefault() :
     + overwrite value (since set to default in constructor)
   + validate_region|shape|shapeMethod|shapeConversion : generic checks for q|t
   + validate_seedRange : generic checks for q|t
   + validate_numberArgumentExcludeRange() : generic check excluding a range
   + validate_id() : checks for line breaks in q|tId
   * parseSequences() :
   * parseSequencesFasta() :
     + idPrefix handling
   - obsolete functions
     - validate_query|target
     - validate_qAccW|L|Constr
     - validate_q|tRegion|Shape|ShapeMethod|ShapeConversion
     - validate_seedQ|TRange
   * constructor() :
     + extended default reset for new arguments
     * q|tAcc* arguments now hidden
     * validation calls refactored
     + model|acc|intLenMax|intLoopMax now general arguments
   * parse() :
     + additional checks for meta arguments with setup of q|t variables
     + usage of .isSet() where appropriate
     + check that outSep is not within id prefix in outMode=C
     + ensure intLenMax >= seedBP

200207 Martin Raden + Fabio Gutmann
 + python/CopomuS.py : Compensatory mutation selector to support interaction
   validation experiments
 + python/copomus/* : utility functions of CopomuS
 * python/README.md : links to dedicated README.md of subfolders

200131 Martin Raden
 * IntaRNA/PredictorMfe2dHeuristicSeedExtension :
   * fillHybridE_left() :
     + extended Zall update (additional save cases considered)


################################################################################
### version 3.1.5
################################################################################  

# IntaRNA :
- bugfix traceback of interactions with seed on right-boundary
- bugfix traceback of seeds with bulges when outNoLP present

################################################################################

200131 Martin Raden
 * IntaRNA/PredictorMfe*SeedExtension* :
   * undo bugfix
   * traceback() :
     + additional check if seed exceeds right interaction boundary
 * IntaRNA/SeedHandler :
   * isFeasibleSeedBasePair() :
     * debug checks obsolete since part of the check
   * updateToNextSeed() :
     * bugfix : right boundary was exclusive (but has to be inclusive)
 * IntaRNA/SeedHandlerMfe :
   * getSeedE()
   * setSeedE()
   * traceBackSeed()
     * now using global indices (offset shift done internally)
   * fillSeed() :
     + additional feasibility test for noLP
   * traceback() :
     * bugfix noLP energy trace : energy was from wrong cells
     * trace small to large gaps (should be faster)
 * test/SeedHandlerMfe :
   + traceback tests (number of bps in traced seeds)

################################################################################
### version 3.1.4
################################################################################  

# IntaRNA
- bugfix generation and tracing of seeds with bulges and no GU ends
- bugfix seed-extension prediction for seeds with bulges
- noLP for seeds with bulges enabled

# R
- `IntaRNA_CSV_p-value.R` script to estimate p-values based on energy values
- `IntaRNA_plotRegions.R` = renaming of former `plotRegions.R`

################################################################################  

200130 Martin Raden
 * IntaRNA/SeedHandlerMfe :
   * bugfix generation and tracing of seeds with bulges and no GU ends
 * IntaRNA/PredictorMfe*SeedExtension* :
   * bugfix enumeration of seeds with bulges
 * bin/CommandLineParseing :
   * error msgs rephrased
   + noLP for seeds with bulges enabled
   + setup noLP for seed constraints via outNoLP
 * IntaRNA/SeedConstraint :
   + isLpAllowed : whether or not lps are allowed in seeds
 * IntaRNA/SeedHandlerMfe :
   + support for noLP constraint
 * test/SeedHandlerMfe :
   + test with lp
   + test no lp (boundary)
   + test no lp (internal)
 * test/*
   * adaptation to SeedConstraint constructor changes

200121 Martin Raden
 + R/IntaRNA_CSV_p-value.R : former addPvalues2csv.R
 + R/IntaRNA_plotRegions.R : former plotRegions.R
 - R/addPvalues2csv.R : renamed
 - R/plotRegions.R : renamed
 * README.md : adapted to renamings
 + R/Makefile.am : install R scripts
 
191115 Martin Raden
 + R/addPvalues2csv.R
 * R/README.md :
   + docu of addPvalue2csv.R

################################################################################
### version 3.1.3
################################################################################  

# IntaRNA
- bugfix latest changes on traceback

################################################################################  

191104 Martin Raden
 * PredictorMfe2dSeedExtension::
   * traceBack() 
     * bugfix tracback (reset of k2 iteration)

191031 Martin Raden
 * using c++11 raw strings where appropriate

################################################################################
### version 3.1.2
################################################################################  

# IntaRNA
- `--outSep` = user-defined column separator for tabular CSV output
- bugfix non-overlapping suboptimal enumeration
- bugfix noLP optimization (missing case of direct left-stack extension)
- CSV output
  - new ensemble energy and partition function output for intra-molecular 
    structures formed by seq1 and seq2 (`Eall1, Eall2, Zall1, Zall2`)
  - new total energy output `Etotal` = (E+Eall1+Eall2) and 
    `EallTotal` = (Eall+Eall1+Eall2)
  - new `RT` output
  - new `bpList` output

# auxiliary R scripts
- plotRegions.py - visualization of sequence regions covered by IntaRNA 
  predictions, similar to the IntaRNA webserver output (thanks to @dgelsin)

################################################################################

191030 Martin Raden
 * IntaRNA/OutputHandlerCsv :
   + RT
   * string2list() :
     + support of '*' encoding to generate full list
 * bin/CommandLineParsing :
   + docu and implementation of '*' outCsvCol behaviour  
 * IntaRNA/InteractionEnergyBasePair :
   + computeIntraEall() : computes Eall1|2 via NussinovHandler
   * getEall1|2() : call computeIntraEall if needed
 * README.md
   + RT CSV col

191029 Martin Raden
 * IntaRNA/PredictorMfe :
   * getNextBest() :
     * bugfix: energy check was applying duplicated ED values 
       (thanks to Jens Georg)
 * IntaRNA/PredictorMfe2dSeedExtension :
 * IntaRNA/PredictorMfe2dHeuristicSeedExtension :
 * IntaRNA/PredictorMfeEns2dSeedExtension :
   * fillHybrid*_left() : 
   * traceBack() : 
     * bugfix: missing case in noLP mode (direct left-stack extension)
 * IntaRNA/AccessibilityVrna :
   + addConstraints() : dedicated function to add constraint to VrnaFoldCompound
   * fillByRNAplfold() : using addConstraints()
 * IntaRNA/InteractionEnergy :
   + getBoltzmannWeight( Z_type ) : conversion from kcal/mol-based energies
   + getEall1|2() : ensemble energy for seq1|2
 * IntaRNA/InteractionEnergyBasePair :
   + getEall1|2() : NOT IMPLEMENTED YET
 * IntaRNA/InteractionEnergyIdxOffset :
   + getEall1|2() : forward to wrapped energy handler
 * IntaRNA/InteractionEnergyVrna :
   + Eall1|2 : ensemble energies for seq1|2
   + getEall1|2() : lazy computation of Eall1|2 using computeIntraEall()
   + computeIntraEall() : Eall* computation via vrna_pf() using the respective
     accessibility constraints
 * IntaRNA/OutputHandlerEnsemble :
   - no output of Zall
   + output of RT, Eall1, Eall2, EallTotal 
 * IntaRNA/OutputHandlerCsv :
   + Eall1, Eall2, EallTotal, Etotal, Zall1, Zall2
 * README.md :
   + docu of new CSV columns (Eall1, Eall2, Zall1, Zall2, Etotal, EallTotal)
   * ensemble output docu updated

191021 Martin Raden
 + R/plotRegions.R :  visualization of sequence regions covered by RRI
 + R/README.md : plotRegions docu
 * README.md : link to R/README.md

191008 Martin Raden
 * IntaRNA/OutputHandlerCsv :
   + bpList output : list of base pairs
 * README.md :
   + bpList 
   + outSep

191008 Martin Raden
 * bin/CommandLineParsing :
   + 'outSep' argument to set column separator for tabular CSV output
   * outSep applied to PredictionTracker*
 * IntaRNA/PredictionTracker* : 
   + explicit output column separator
 * IntaRNA/OutputHandlerCsv :
   * needsZall() :
   * needBPs() :
     - colSep argument : obsolete
  
191007 Martin Raden
 * bin/CommandLineParsing :
   * bugfix accNoLP and accNoGUend checks for energy!=V
 * IntaRNA/NussinovHandler :
   * getQ() :
     * fix: return 1 if (i==j)
 * IntaRNA/AccessibilityVrna :
   * construction() : 
     + bugfix: missing ED initialization for short sequences
 * tests/AccessibilityBasePair :
   + short sequence test
 + tests/AccessibilityVrna  :
   + short sequence test to validate ED initialization

################################################################################
### version 3.1.1 
################################################################################  

# IntaRNA
- base pairs details only computed if needed for output (speedup for large -n)
- predefined parameter sets for loading (Turner04, Turner99, Andronescu07)
- 'tRegion' and 'qRegion' now available for and applied to multi-sequence input

################################################################################  
  
190924 Martin Raden
 * IntaRNA/general :
   - numStringLength() : obsolete
 * IntaRNA/OutputHandlerText : 
   + explicit '+' for positive indices if relative indexing enabled
 * bin/CommandLineParsing :
   * validate_indexRangeList() :
     + sequence id in error message
   * parseRegion() :
     + range check per sequence with sequence-specific error message
     - single sequence input condition removed : ranges applied on all sequences
 * IntaRNA/AccessibilityVrna :
   * construction() : 
     + bugfix: only call VRNA routines if seq.size() > 4
  
190923 Martin Raden
 + m4/m4-ax_boost_iostreams.m4
 + m4/m4-ax_check_zlib.m4
 * configure.ac :
   * ViennaRNA version >= 2.4.14
   + check for boost::iostreams library
   + check for zlib library
 * IntaRNA/OutputConstraint :
   + needBPs : whether or not traceback of interaction base pairs is needed
 * IntaRNA/PredictorMfe :
   * reportOptima() :
     + traceback now optional based on OutputConstraint::needBPs
 * IntaRNA/OutputHandlerCsv :
   + needBPs() : checks whether or not base pair information is needed for the
     CSV columns to print
 * IntaRNA/VrnaHandler :
   + Turner99 : const string representing RNA-Turner-1999 energy file
   + Turner04 : const string representing RNA-Turner-2004 energy file
   + Andronescu07 : const string representing RNA-Andronescu-2007 energy file
   * constructor() :
     * Turner04 = default parameter set
     + load of Turner99/04 and Andronescu07 parameters via new VRNA-2.4.14 API
 * bin/CommandLineParsing :
   + outNeedsBPs : whether or not output requires interaction traceback
   * energyVRNA :
     + docu of Turner99/04 and Andronescu07 predefined models for loading
     + default value = Turner04
   * IntaRNA1 :
     + default energyVRNA = Turner99
   * getOutputHandler() :
     + setup of outNeedsBPs
 * IntaRNA/Interaction :
   * dotBar : indexing based on RnaSequence::getInOutIndex()
 * IntaRNA/OutputHandlerText :
   * add() : 
     * bugfix: isGU() test was using non-transformed bp indices
 * README.md :
   + docu of predefined parameter sets 

################################################################################
### version 3.1.0.2  
################################################################################  

190920 Martin Raden :
 * IntaRNA/PredictorMfe :
   * getNextBest() :
     * ensure deterministic output for degenerate bestE interactions (left most)
 * tests updated

################################################################################
### version 3.1.0.1
################################################################################  

# IntaRNA
+ Zall (estimates) and derived output now available for all prediction models
+ outMode=E : report of ensemble information only
* ensemble-based prediction revised
* seed-extension-based prediction revised
+ --seedNoGUend : prohibit GU base pairs at seed ends
+ outNoLP support in all prediction modes
* FASTA input: sequence id = prefix up to first whitespace from FASTA header
* new default for outOverlap = B: overlapping output enabled (was Q)
+ gzip-compressed file input/output if file name ends in '.gz'
+ --outNoGUend : omit interactions with GU-helix ends
+ --accNoLP : exclude lonely base pair in accessibility computation (energy==V) 
+ --accNoGUend : exclude GU-helix ends in accessibility computation (energy==V)
* outMode=D : subsequence range now "START..END" instead of "START--END"
+ --q|tIdxPos0 : user defined sequence indexing for input parsing and output

# package
+ (optional) intarnapvalue python3 module to compute p-values of IntaRNA mfe
  predictions based on sequence shuffling and distribution fitting 

# API
* OutputHandler requires an OutputConstraint instances in construction
* Predictor objects now access OutputConstraint instances via the OutputHandler;
  this simplifies the signature of the member functions
* OutputConstraint now guides whether or not Zall is required for output

################################################################################  

190919 Martin Raden :
 * bin/CommandLineParsing :
   * argument validation revised

190918 Martin Raden :
 * IntaRNA/general :
   * numStringLength() : now template function
 * IntaRNA/RnaSequence :
   + idxPos0 : input/output index of first position
   + getInOutIndex() : converts internal to in/out index
   + getIndex() : converts in/out index to internal indexing
   + getReversedIndex() : backward internal index
 * IntaRNA/AccessibilityConstraint :
   * construction from string now based on RnaSequence::getIndex()
   * regionIndexList : allows for negative input indices
 * IntaRNA/IndexRange :
   + regexString : static string encoding of regex : allows for neg. in indices
   * regex : init based on regexString
   * construction from string now based on RnaSequence::getIndex()
   * fromString() now based on RnaSequence::getIndex()
 * IntaRNA/IndexRangeList :
   * regexString : now based on IndexRange::regexString
   * construction from string now based on RnaSequence::getIndex()
   * fromString() now based on RnaSequence::getIndex()
 * IntaRNA/InteractionEnergy :
   * getE_contributions() 
     * simplification using class members getIndex*()
 * IntaRNA/OutputHandlerCsv :
 * IntaRNA/OutputHandlerText :
   * add() :
     * index shifting for output based on RnaSequence::getInOutIndex()
 * IntaRNA/OutputHandlerEnsemble :
   - outputDone : obsolete
 * IntaRNA/PredictionTrackerPairMinE :
 * IntaRNA/PredictionTrackerSpotProbAll :
   * writeData() :
     * index shifting for output based on RnaSequence::getInOutIndex()
 * IntaRNA/PredictionTrackerProfileMinE :
 * IntaRNA/PredictionTrackerProfileSpotProb :
   * writeProfile() :
     * index shifting for output based on RnaSequence::getInOutIndex()
 * IntaRNA/PredictionTrackerSpotProb :
   * str_spot : allows for negative input indices
   * destructor() :
     * index shifting for output based on RnaSequence::getInOutIndex()
     * "0&0" output changed to "idxPos0-1" to be more general
   * Spot : 
     - construction from string
   + fromString() : creates a Spot from string using RnaSequence::getIndex()
 * IntaRNA/ReverseAccessibility :
   * getReversedIndex() : now based on RnaSequence::getReversedIndex()
 * IntaRNA/SeedHandlerExplicit :
   * checkSeedEncoding() : regex now allows for neg. input indices
   * SeedData :
     * construction from string now based on RnaSequence::getIndex()
 * tests :
   * adaptation of negative index input tests (now valid)
   * RnaSequence :
     + test for getInOutIndex() + getIndex()
 * bin/CommandLineParsing :
   + q|tIdxPos0 : start of indexing for query|target
   * validate_indexRangeList(.., RnaSequence) : to get index shifting
 * README.md :
   + documentation of q|tIdxPos0 effects
   + q|tIdxPos0 sequence indexing section

190913 Martin Raden :
 * configure.ac :
   * LIBS exended
 * IntaRNA/general :
   * new|deleteOutStream :
     + gzip-compression support enabled
     * fstream replaced with boost::iostreams::filtering_ostream
   + new|deleteInStream : analog to outstream handling to enable gzip support
 * IntaRNA/InteractionEnergy :
   - getBestE*() : obsolete
   * isValidInternalLoop() : now public
   + internalLoopGU : whether or not GU is allowed in non-stacked interior loops
   + isInternalLoopGUallowed() : returns internalLoopGU
   * isValidInternalLoop() :
     + check internalLoopGU constraint
 * IntaRNA/Predictor :
   * updateZall() :
     + bugfix: additional check for outConstraint.needZall
 * IntaRNA/PredictorMfe :
   - min*Energy : obsolete
 * IntaRNA/PredictorMfe2dHeuristicSeed :
   * fillHybridE() :
     * bugfix Zall update : seed-only also if !noLP
 * IntaRNA/InteractionEnergyIdxOffset : 
   + bugfix: adding missing offset-based overwrites
 * IntaRNA/OutputConstraint :
   + noGUend : whether or not GU is allowed at interaction ends
 * bin/CommandLineParsing :
   * input file handling replaced with new|deleteInStream
   + outNoGUend : whether or not GU is allowed at interaction and helix ends
   * IntaRNAsTar :
     + outNoGUend
     + outNoLP
 * README.md :
   + gzip compression docu
   + outNoGUend docu
   * IntaRNAsTar docu
 * tests :
   + outNoGUend

190912 Martin Raden :
 + python/intarnapvalue : p-value computation for IntaRNA energy scores
   (thanks to Fabio Gutmann)
 * configure :
   + check for python >= 3.6
   + optional enabling of intarnapvalue installation (disabled by default)
 * tests/data/Makefile.am :
   + distribution of all tests ensured

190911 Martin Raden :
 * IntaRNA/SeedHandler :
   * isFeasibleSeedBasePair() : now virtual and public
 * IntaRNA/SeedHandlerIdxOffset :
   + addSeeds() 
   + isFeasibleSeedBasePair()
   + areLoopOverlapping()
 * IntaRNA/PredictorMfe2dSeedExtension :
   * traceBack() :
     * trace order optimized
 * IntaRNA/PredictorMfeEns2dSeedExtensionHeuristic :
   * fill*() : reduced to updateZ() calls to reduce code redundancy 

190910 Martin Raden :
 * IntaRNA/PredictorMfe2d*SeedExtension :
   + noLP support
 + tests for noLP seed extension (model=X)
 * bin/CommandLineParsing :
   * parseSequencesFasta() :
     * name = FASTA header prefix up to first whitespace
 * IntaRNA/PredictorMfe :
   + HashIdx2E : hash for base pair to BestInteractionE objects
   + mfe4leftEnd : (optional) hash to store best interaction for left boundary;
     filled by updateMfe4leftEnd()
   + updateMfe4leftEnd() : updates mfe4leftEnd
   * getNextBest() :
     + default implementation based on mfe4leftEnd
   * updateOptima() :
     + calls updateMfe4leftEnd() if outconstraint.reportOverlap != BOTH
   * exact + seed-extension predictors :
     - getNextBest() dummy functions : obsolete
     - outConstraint.reportOverlap != BOTH error : obsolete
   * heuristic non-seed-extension predictors :
     + updateMfe4leftEnd() : dummy function to avoid mfe4leftEnd updates; since
       local getNextBest() is based on local data member 
 + tests for outOverlap=NQTB (model=X)

190909 Martin Raden :
 * Makefile.am :
   - explicit personality list for link generation; replaced by
   + automatic link generation for each personality
 * IntaRNA/OutputHandlerCsv : 
   + 'w' = Boltzmann weight output
 * IntaRNA/Predictor :
   + updateZall() : updates Zall with a given interaction energy
 * IntaRNA/PredictorMfe :
   * updateOptima() :
     + calls updateZall() if needed
 * IntaRNA/PredictorMfe2dHeuristic :
   + update of Zall
 * IntaRNA/PredictorMfe2dHeuristicSeed :
   * bugfix Zall update
 * IntaRNA/PredictorMfe2dSeed :
   * bugfix updateOptima() call
 * IntaRNA/PredictorMfeEns2dHeuristic :
   * bugfix total energy computation
   + updateZ() call corrected
 * IntaRNA/PredictorMfeEns2dSeedHeuristic :
   * updateZ() call order changed to reduce for-loop iterations
 * README.md :
   + w CVS output documentation
   + E_add CVS output documentation
   + RT(37Celsius)=0.6163173043012 docu

190906 Martin Raden :
 * IntaRNA/Interaction
   + Boundary : 4-tuple of indices to store boundaries of an interaction site
 * tests/runIntaRNA.sh :
   * tests are run for all data/*.parameter files (no explicit listing anymore)
 + tests/*-modelX|S : explicitly testing model=X or =S
 * IntaRNA/PredictorMfe2dHeuristicSeed :
   * fillHybridE() : updateOptima() now called for all tested cases to reduce
     heuristic bias in subopt listing
 * IntaRNA/PredictorMfeEns :
   * renaming Z_partitions to Z_partition
   + updateOptimaUsingZ() : calls updateOptima() for all Z_partition entries
   + reportOptima() : calls updateOptimaUsingZ() and super class reportOptima()
   + traceBack() : central traceback for all ensemble-based predictors
   + BestInteraction template class
   + BestInteractionE|Z typedefs
   * Z_partition now hash of Interaction::Boundary keys that map to Z_type
   - checkKeyBoundaries() : obsolete
 * IntaRNA/PredictorMfeEns2d :
   - traceBack()
   * predict() :
     - explicit updateOptima() calls
 * IntaRNA/PredictorMfeEns2dHeuristic :
   * predict() :
     - explicit updateOptima() calls
   * fillHybridZ() :
     + explicit updateZ() calls for all recursion cases to reduce heuristic bias
   - BestInteraction : replaced by PredictorMfe::BestInteraction
 * IntaRNA/PredictorMfe2dSeedExtension :
   - updateOptima() : obsolete
   * predict() :
     + explicit j boundary computation to reduce considered index tuples when
       updating optima
 * IntaRNA/PredictorMfe2dSeedExtensionRIblast :
   - updateOptima() : obsolete
 * IntaRNA/PredictorMfe2dHeuristicSeedExtension :
   * rename updateOptima() to updateOptimalRightExt()
   + updateOptima() called for all tested right extensions
   + updateOptima() called for all tested seed+leftExtension
   + extended search range for seed+leftExtension (so far limited by jOptRight)
 * IntaRNA/PredictorMfeEns2dSeedExtension :
   - explicit updateOptima() calls
   * traceBack() : reduced to super class traceBack() call seed addition
 * IntaRNA/PredictorMfeEns2dHeuristicSeedExtension :
   - explicit updateOptima() calls
   + extended search range for seed+leftExtension (so far limited by jOptRight)
   + updateZ() called for all tested right extensions
   + updateZ() called for all tested seed+leftExtension
 * bin/CommandLineParsing :
   + check for outOverlap=B if window-based computation with subopt output (n>1)
 * IntaRNA/PredictorMfe2dHeuristic :
   - BestInteraction : replaced by PredictorMfe::BestInteraction
 * IntaRNA/SeedConstraint :
   + bpGUendAllowed + is*() : whether or not GU is allowed at seed ends
 * IntaRNA/SeedHandler :
   * isFeasibleBasePair() :
     + atSeedEnd argument: to enable seed end check if needed
     + explicit GU-end checks if atSeedEnd==true 
 * IntaRNA/SeedHandlerMfe :
   + check for GU ends
 * IntaRNA/SeedHandlerNoBulge :
   * storeSeed() : 
     + check for GU ends
 * README.md :
   + seedNoGUend documentation
 
190905 Martin Raden :
 * IntaRNA/PredictorMfe2dHeuristicSeedExtension :
   + updateOptima() also for left+seed called (less heuristic bias)
 * IntaRNA/PredictorMfeEns* : ensemble-mfe prediction revised

190830 Martin Raden :
 + IntaRNA/OutputHandlerEnsemble : reports ensemble information only

190828 Martin Raden :
 * Makefile.am :
   + show test log if tests failed
 * IntaRNA/OutputHandler* :
   * bugfix: outConstraint was reference instead of full object

190827 Martin Raden :
 * IntaRNA/OutputConstraint :
   + needZall : whether or not Zall is needed for output generation
 * IntaRNA/OutputHandler* :
   + outConstraint : OutputConstraint to be heeded
   * constructor : required OutputConstraint object
   * add() : outConstraint parameter obsolete
 * IntaRNA/OutputHandlerCsv :
   + needsZall() : checks whether or not the requested column types require Zall
 * IntaRNA/Predictor* :
   * outConstraint parameter in member functions obsolete
 * bin/CommandLineParsing :
   + outNeedsZall : flag to check whether or not Zall is needed for output
 * IntaRNA/Predictor :
   + Zall : overall partition function
   * updateOptima( ..., incrementZall = true ) : Zall is incremented
     if output.outConstraint.needZall == true
   + getZall() : access to overall partition function
   + incrementZall() : increments Zall (+ overflow check in debug mode)
   * reportOptima() :
     + calls output.incrementZ()
   * initOptima() :
     + init of Zall if needed
 * IntaRNA/PredictorMfeEns :
   - overallZ : replaced by Predictor::Zall
   - reportOptima() : obsolete
 * IntaRNA/PredictorMfeEns* :
   * call of updateOptima( ..., incrementZall = false ) to avoid duplicated Zall
     updates
   * initZ() : - no init of Zall


################################################################################  
##### version 3.0.0
################################################################################  

+ personality
  + IntaRNA1,2,3
  + IntaRNAexact
  + IntaRNAhelix
  + IntaRNAduplex
+ outNoLP
+ outCsvSort
+ output of all putative seeds (unless outMode=N or outBestSeedOnly)
+ seed bps are highlighted with '+' in text output mode (minimal + detailed)
* model=X : new default = seed-extension-based prediction
+ model=P
+ support for multi-precision Z (via GNU quadmath library) : configure
+ mode=S
+ energyNoDangles
+ parameterFile
+ seedNoGU
+ seedMaxEhybrid
+ helixMinPu (replaces helixMaxED)
+ helixFullE : replaces helixNoED
+ model=B (replaces L)
* reporting best (mfe) seed per interaction
* outMode=C
  + sorting
  + columns Eall, Zall, P_E
  * marker "NAN" for not available values
* outMode=D
  * index ranges without whitespaces, i.e. "1--5" instead of "1 -- 5"
  + highlighting of all bps part of putative seeds
  * gaped loop positions now represented by '-'
* E_type now int (was float)
- IntaRNA v1 output support not continued

################################################################################  

190809 Martin Raden :
 * IntaRNA/general :
   * deleteOutStream() :
     * now reference to stream pointer to ensure update of original pointer
   * E_INF = INTMAX / 8 (to ensure correct boundary checks in predictors)
 * IntaRNA/OutputStreamHandlerSortedCsv :
   * destructor() :
     * BUGFIX: double deletion of outstream (fixes #157)
 * IntaRNA/PredictorMfe2dSeed :
   * traceback() :
     * BUGFIX noLP traceback (fixes #153)
 * IntaRNA/OutputHandlerText :
   * add():
     * non-existing positions in loops are now represented by '-' symbol (#155)
     * always highlight seed bps with '+'
 * bin/CommandLineParsing :
   + resetParamDefault() : resets a parameter's default value and reports VLOG
   + IntaRNA1 personality: defaults similar to v1
   + IntaRNA2 personality: old v2 defaults
   + IntaRNA3 personality: IntaRNA alias
 * tests/data/*.parameter :
   * adapt to new default model=X (was S)
 + tests/data/Makefile.am : add test data to distribution

190808 Martin Raden :
 * renaming personality IntaRNAblock to IntaRNAhelix (since used in resp. paper)

190716 Martin Raden :
 * bin/CommandLineParsing :
   * IntaRNAexact :
     + model=X
     + exact accessibility computation (*AccW = *AccL = 0)

190716 Martin Raden :
 * IntaRNA/Accessibility :
   * decomposeByMaxED( ranges ) : 
     + additional parameter for minimal range length
 * bin/CommandLineParsing :
   - high memory warning disabled
   + set seedBP for minimal range length of range decomposition
 * IntaRNA/IndexRange :
   * overlappingWindows :
     * bugfix: check for non-ascending was not using "!isAscending()"

190607 Martin Raden :
 * IntaRNA/general :
   * deleteOutputStream() :
     - no warning if not a file stream (to support temporary stringstreams)
 * IntaRNA/OutputHandlerCsv :
   + colTypeNumericSort : list of types that need numerical sorting
   * notAvailable : now "NAN" (was "NA")
   - colOrder : renamed to columns
   + columns (former colOrder)
 + IntaRNA/OutputStreamHandler : generic handler to wrap the output stream
 + IntaRNA/OutputStreamHandlerSortedCsv : OSH that stores output in temporary
   stringstream to be finally sorted by the values of a given column
 * bin/CommandLineParsing :
   - outStream : replaced by outStreamHandler
   + outStreamHandler : central OutputStreamHandler object
   + outCsvColSep : CSV column separator
   + outCsvLstSep : CSV value-list separator 
   + outCsvSort : optional parameter to define the CSV column for sorting
   * destructor() :
     * outStream deletion
   * parse() :
     + parsing and setup CSV sorting
   - getOutputStream() : replaced by outStreamHandler->getOutStream()
 * README.md :
   + outCsvSort documentation
 - perl/IntaRNA_1ui.pl : v1 support ends   
 - perl/IntaRNA_up_1ui.pl : v1 support ends   

190606 Martin Raden :
 * IntaRNA/OutputHandlerText :
 * IntaRNA/OutputHandlerCsv :
   * add() :
     * bugfix: Pu values were truncated like energies 
 * IntaRNA/HelixHandlerNoBulgeMax :
   * fillHelixSeed() :
     * bugfix: leadingE computation was shifted by one index to the right
 * bin/CommandLineParsing :
   + output alias information [seq1|seq2] added to target|query parameter info
   + outCsvSort parameter to trigger sorted CSV output
   * getOutputHandler() :
     + returns OutputHandlerCsvSorted if outCsvSort non-empty
 * README.md 
   * example output updated
   + output alias information [seq1|seq2] added to detailed mode output

190605 Martin Raden :
 * IntaRNA/HelixHandlerNoBulgeMax :
   * fillHelixSeed() :
     * bugfix: complementarity check was using (i2,i2)

190603 Martin Raden :
 * IntaRNA/PredictorMfeEns2d :
 * IntaRNA/PredictorMfeEns2dHeuristic :
   * predict() :
     * bugfix: noLP handling corrected
 * IntaRNA/Predictor*SeedExtension* :
   + warning that noLP not implemented yet
 * renaming IntaRNA/PredictorMfeEns2dSeedExtensionRiBlast to
   IntaRNA/PredictorMfeEns2dSeedExtensionRIblast

190523 Frank Gelhausen + Martin Raden :
 - IntaRNA/PredictorMfe4d : obsolete and was for testing only
 - IntaRNA/PredictorMfe4dSeed : obsolete and was for testing only
 + IntaRNA/PredictorMfe2dSeedExtension : exact mfe via SE
 + IntaRNA/PredictorMfe2dSeedExtensionRiBlast : RIblast-like SE mfe
 + IntaRNA/PredictorMfe2dHeuristicSeedExtension : heuristic SE mfe
 + IntaRNA/PredictorMfeEns2d : exact ensemble mfe
 + IntaRNA/PredictorMfeEns2dHeuristic : heuristic ensemble mfe
 + IntaRNA/PredictorMfeEns2dSeedExtension : exact ensemble mfe via SE
 + IntaRNA/PredictorMfeEns2dSeedExtensionRiBlast : RIblast-like SE ensemble mfe
 + IntaRNA/PredictorMfeEns2dHeuristicSeedExtension : heuristic SE ensemble mfe
 * configure.ac :
   + support for multi-precision Z (via GNU quadmath library)
 * IntaRNA/general :
   + support for multi-precision Z (via GNU quadmath library)
   + Z_log & Z_exp : placeholder to call correct function for current Z_type
   - E_precisionEpsilon : obsolete
 * IntaRNA/OutputHandlerText :
   * range encoding now without whitespace, i.e. "i--j" instead of "i -- j"
 * IntaRNA/PredictorMfeEns :
   - getHybridZ() : replaced by getOverallZ()
   + getOverallZ()
   + checkKeyBoundaries() : to check if site-specific Z hashing is possible
   + ZPartition : container to store and update site-specific Z values
   + Z_partitions : member to store site-specific Z values
   * updateZ() :
     + check for partZ == 0 or INF
     + update of Z_partitions

190523 Martin Raden :
 - IntaRNA/PredictorMaxProb : obsolete (will be replaced soon)
 + IntaRNAens personality (model=P)
 * IntaRNA/IndexRangeList :
   * constructor(stringEncoding, +allowOverlap) : explicit init of allowOverlap
 * IntaRNA/OutputHandler :
   + Z + getZ() : overall partition function (if updated by predictor)
   + incrementZ() : update of Z (assuming independent interaction sets)
 * IntaRNA/OutputHandlerText :
   + print ensemble energy if computable
 * IntaRNA/OutputHandlerCsv :
   * column docu corrected for E and seedE
   + Eall, Zall, P_E
   + notAvailable : string to mark not available values
   * add() :
     * replace signalling_NaN() with notAvailable
     + handling of Eall, Zall, P_E
 * IntaRNA/PredictorMfe :
   * predict() :
     + output.add() calls now omp_critical (since multiple predictors can call
       in parallel)
 * IntaRNA/PredictorMfeEns :
   - overallZhybrid : replaced by overallZ
   + overallZ
   + reportOptima() : calls output.incrementZ(overallZ) to report overall Z
   * initZ() :
     - no call of initOptima()
 + IntaRNA/PredictorMfeEnsSeedOnly : ensemble-based seed-only predictions
 * bin/IntaRNA :
   + transfer accumulated Z information from bestInteractions to final output
 * bin/CommandLineParsing :
   - remove documentation of (non-implemented) BED-file support from q|tRegion
   * updating model=P docu
   * parse() :
     + check for model=P that no window-based prediction (wrong Z aggregation)
   * validateRegion() :
     + explicit non-overlapping list check
   * parseRegion() :
     * cleanup
 * README.md :
   + IntaRNAens
   + model=P
   + CSV output : Eall, Zall, P_E

190522 Martin Raden :
 * renaming personality IntaRNAup to IntaRNAexact
 * renaming personality IntaRNAhelix to IntaRNAblock
 + personality IntaRNAsTar : optimized parameters for sRNA-target prediction  

190521 Martin Raden :
 * IntaRNA/AccessibilityVrna :
   - calc_ensemble_free_energy() :
   - getPfScale() :
   - fillByConstraints() :
   - fillByRNAup() : obsolete and not conform to new VRNA API
 * bin/CommandLineParsing :
   + check for noSeed+mode=S and throw error

190418 Martin Raden :
 + IntaRNA/PredictorMfeSeedOnly : predicts only seed interactions
 * bin/CommandLineParsing :
   + mode=S : predicting seeds only
   + personality=IntaRNAseed : enabling mode=S

190414 Martin Raden :
 * bin/CommandLineParsing :
   * bugfix: helix constraints checks only done for model = B
   + mode : renamed predMode
   - predMode : renamed to mode
   * NumberParameter :
   * CharParameter :
     * def value now non-constant to be changeable in constructor 
   * constructor() :
     - variable init now non-dynamic
     + explicit en-bloc default update for each personality
     + report of default changes via VLOG

190412 Martin Raden :
 * IntaRNA/Interaction :
   * dotBracket() :
     * bugfix: full string generation produced wrong tails

190411 Martin Raden :
 * bin/CommandLineParsing :
   + link to github pages for further documentation

190409 Martin Raden :
 * IntaRNA/InteractionEnergy :
   + energyWithDangles : whether or not dangling end contributions are to be 
     considered for the overall energy computation
   * getE() :
   * getE_contributions() :
   * getE_multi() :
     + consideration of energyWithDangles parameter
   * getE_multi() :
     + bugfix: helix-end terms were missing
 * IntaRNA/InteractionEnergy* :
   * constructor() :
     + energyWithDangles handling
 * bin/CommandLineParsing :
   + energyNoDangles : whether or not dangling end terms are considered
   * getEnergyHandler() :
     * bugfix : InteractionEnergyBP constructor call

190404 Martin Raden :
 * IntaRNA/InteractionEnergy :
   * getE() :
     + additional sanity checks to avoid energy range overflows

190403 Martin Raden :
 * IntaRNA/PredictorMfe :
   * updateOptima() :
     + bugfix: explicit handling of energies >= E_MAX

190320 Martin Raden :
 * IntaRNA/PredictorMfe2dHeuristic* :
   + outNoLP support
 * tests/data :
   + tests with and without noLP
 * .travis.yml : 
   - viennarna not available via ppa anymore
   + build IntaRNA via conda
 * tests/Makefile.am :
   * switch to parallel test harness to define shell script test driver

190319 Martin Raden :
 + tests/runIntaRNA.sh : runs tests using the IntaRNA binary
 + tests/data :
   + noLP-exact 
   + noLP-exact-seed
 * IntaRNA/Predictor* :
   * predict() :
     + "outNoLP not implemented" warnings where needed
 * bin/CommandLineParsing :
   * getSeedHandler() :
     + "outNoLP not implemented" warning for SeedHandlerMfe
 * README.md :
   + outNoLP

190318 Martin Raden :
 * IntaRNA/OutputConstraint :
   + noLP : whether or not lonely (non-stacked) interaction bps are allowed
   * constructor( + noLP )
 * IntaRNA/InteractionRange :
   * energy init = E_INF (was signaling_NaN())
 * IntaRNA/Interaction :
   * energy init = E_INF (was signaling_NaN())
   * IntaRNA/Interaction::Seed :
     + operator == ()
   + operator == ()
     + check for second base pair
 * IntaRNA/PredictorMfe :
   * updateOptima() :
   * reportOptima() :
     * bugfix of degenerate energy handling + simplification
 * IntaRNA/PredictorMfe2d :
 * IntaRNA/PredictorMfe2dSeed :
   + support for OutConstraint::noLP
   * fillHybridE( + callUpdateOptima ) : updating optima now optional
   * traceBack() :
     * bugfix : call of fillHybridE( callUpdateOptima = false ) to avoid 
       redundant updateOptima() calls
   + support for OutConstraint::noLP
 * bin/CommandLineParsing :
   + outNoLP : whether or not lonely (non-stacked) interaction bps are allowed

190315 Martin Raden :
 * IntaRNA/Accessibility :
   + decomposeByMaxED() : replaces decomposeByMinPu()
   - decomposeByMinPu() : renamed to decomposeByMaxED()
 * IntaRNA/InteractionEnergy :
   + energyAdd + get*() : shift of the energy spectrum to be applied in getE()
   * constructor( + energyAdd )
   * getE() :
     + energyAdd
   * EnergyContributions :
     + energyAdd member
   * getE_contributions() :
     + energyAdd
 * IntaRNA/InteractionEnergy* :
   * constructor( + energyAdd )
 * SeedHandler :
   + areLoopOverlapping() : tests whether or not the last x and the first x
     loops of two seeds are formed by identical base pairs
 - IntaRNA/OutputHandlerRangeOnly : obsolete
 * OutputHandler* :
   * add( + outConstraint )
   - add( InteractionRange ) : obsolete
 * IntaRNA/OutputHandlerText :
 * IntaRNA/OutputHandlerCsv :
   - outConstraint : now provided as argument to add()
   * constructor( - outConstraint )
 - IntaRNA/OutputHandlerRangeOnly : obsolete
 * IntaRNA/PredictorMfe :
 * IntaRNA/PredictorMaxProb :
   * adaptation to OutputHandler::add() change
 * bin/CommandLineParsing :
   + calling decomposeByMaxED() only if outMinPu > 0
   + energyAdd parameter + validate*() to shift energy spectrum
   * getInteractionEnergy() : 
     + set energyAdd
 * BOOST_FOREACH replaced by c++11 for loop syntax

190313 Martin Raden :
 * IntaRNA/IndexRangeList :
   + allowOverlapping member
   + isAllowingOverlap() 
   * constructor( + allowOverlapping )
   + reverseInplace() : renamed non-const reverse method to make intention clear
   - reverse() : replaced by reverseInplace()
   * ostream operator now friend function
   * covers() :
   * overlaps() :
   * insert() :
     + support for non-overlapping ranges added
 * IntaRNA/Interaction : 
   * Seed :
     + constructor() to enable default and explicit initialization
     + < operator to enable automated (energy-based) sorting
   + SeedSet : (energy-sorted) set of seeds
   * seed member now SeedSet* (was Seed*) to support storage of multiple seeds
   + getSeedRanges*() : provides an (overlapping) IndexRangeList of all seeds
     stored within the interaction for the respective sequence
   * ostream operator( BasePair ) now friend function
 * IntaRNA/SeedHandler :
   * traceBackSeed() now const function
   + isSeedBound() : to check whether or not base pair marks a left-side seed
     boundary
   + addSeeds() : adds all seeds that overlap with base pairs of a given
     interaction to its internal seed list
 * IntaRNA/SeedHandler* :
   + isSeedBound() 
   * traceBackSeed() const
 * IntaRNA/SeedHandlerExplicit :
   + getSeedMaxBP() : provides the maximal number of base pairs within one of
     the explicit seed encoded by the provided string
 * IntaRNA/SeedHandlerMfe :
   * getSeedE() const
 * IntaRNA/Predictor*Seed :
   + using SeedHandler::isSeedBound()
   + calling seedHandler.addSeeds(interaction) for each interaction
 * IntaRNA/HelixHandler* :
   + using SeedHandler::isSeedBound()
 * IntaRNA/OutputConstraint :
   + bestSeedOnly : whether or not only the mfe seed is to be printed
   * constructor( + bestSeedOnly )
 * IntaRNA/OutputHandler* :
   * adaptation to Interaction.seed change (SeedSet instead of Seed object)
 * IntaRNA/OutputHandlerCsv :
 * IntaRNA/OutputHandlerText :
   + outConstraint : output constraint also used for predictors
   * constructor( + outputConstraint )
   * add() :
     + support for multi seed output
     + support for mfe seed output based on OutConstraint::bestSeedOnly
 * IntaRNA/OutputHandlerCsv :
   + listSep : separator of multi value entries within one CSV column
   * constructor( + listSep )
   * list2string() : 
     + sep argument to explicitly specify separator to be used
   * getHeader() : now based on list2string()
 * IntaRNA/OutputHandlerText :
   * add() :
     + seed base pairs now represented by '+' (in detailed output mode)
     * GU base pair check via InteractionEnergy::isGU()
 * tests : adaptation to changes
 * tests/SeedHandlerNoBulge :
   + test isSeedBound()
 + tests/runApiTests.cpp : renamed runTests.cpp
 - tests/runTests.cpp : replaced by runApiTests.cpp
 * IntaRNA/AccessibilityVrna :
   * fillByRNAplfold() :
     + throws error if vrna_probs_window() reports problems (zero return value)
 * bin/CommandLineParsing :
   + mode : renamed predMode
   - predMode : replaced by mode
   + outBestSeedOnly : whether or not to print only mfe seed information
   * check if helixMaxBP >= seedBP now handle explicit seed input
   * check for minimal sequence length based on seed now only if seeds computed
 * README.md :
   + outBestSeedOnly parameter 
   + '+' seed bps in detailed text output
   * restructuring of section order + new main sections defined

190312 Martin Raden :
 * IntaRNA/Accessibility* :
   + verbose output how accessibility is computed
 * bin/CommandLineParsing :
   + Personality + get/Name() : enum of IntaRNA's known multiple personalities
   + getPersonality() : checks how IntaRNA was called to react accordingly...
   + personality argument and member
   * construction() :
     + takes now Personality instead of callName
     + default setup based on selected personality
 * Makefile.am :
   + revealing personalities IntaRNAup, IntaRNAduplex and IntaRNAhelix as links 
     in install directory
 * README.md :
   + multiple personalities and their use cases

190311 Martin Raden :
 * configure.ac :
   + AC_PROG_LN_S : check for symboling linking
 * Makefile.am :
   + create IntaRNA alias as link to setup alternative defaults and behavior 
 * bin/CommandLineParsing :
   * construction( callName ) : takes argv[0] (== name of the called binary) to
     setup default arguments and behavior of the respective IntaRNA alias

190310 Martin Raden :
 + IntaRNA/PredictorMfeEns : base class for ensemble-based predictors
   + overallZhybrid : overall hybridization partition function since last 
     predict() call
   + initZ()
   + updateZ()
   + getHybridZ() : access to the overall hybridization partition function

190308 Martin Raden :
 * bin/CommandLineParsing :
   + --parameterFile argument (specify additional parameters via config file);
     command line arguments take precedence over config file parameters
   + configFileName : file from where to read parameters from
   * bool flags now set via boost not via explicit count check
   + bool flags can now be explicitly set with value , eg --noSeed=false, eg
     to overwrite arguments from config file  
 + doc/figures/freepik-*.zip : original icon downloads from freepik.com
 + doc/figures/icon-up.* : up arrow to be used for overview navigation
 * README.md :
   + parameterFile docu
   + direct links to overview (table of content) after each subsection

190307 Martin Raden :
 * IntaRNA/SeedConstraint :
   + bpGUallowed : whether or not GU base pairs are allowed within seeds
   + isGUallowed() : returns bpGUallowed
   + maxEhybrid : maximal hybridization energy
   + getMaxEhybrid()
 * IntaRNA/RnaSequence :
   + bpGUcodes : base pair codes for GU and UG
   + isGU() : tests whether or not two positions can form a GU base pair
 * IntaRNA/InteractionEnergy :
   + isGU() : tests whether or not two positions can form a GU base pair
 * IntaRNA/SeedHandler :
   * isFeasibleSeedBasePair() :
     + check if GU base pair are allowed or not
 * IntaRNA/HelixHandlerNoBulgeMax :
   * fillHelixSeed() : 
     * now based on seed iteration to speed up
 * CommandLineParsing :
   + helixMinPu : minimal unpaired probability (per sequence) of considered
     helices (to make interface analogously to seed constraints)
   - helixMaxED : replaced by helixMinPu
   + seedNoGU : if present, GU base pairs are not allowed within seeds
   + seedMaxEhybrid : maximal hybridization energy (excl.) for seeds
 * README.md :
   + seedNoGU
   + seedMaxEhybrid
   + helixMinPu
   - helixMaxED
   * figure links corrected (to be working within github pages) 

190306 Martin Raden :
 * README.md :
   + helix constraints documented
 * IntaRNA/HelixHandlerUnpaired :
   * adaptation to HelixConstraint::evalFullE() semantics
 * bin/CommandLineParsing :
   + helixFullE : replaces helixNoED
   - helixNoED : replaced by helixFullE
   * model : renamed 'L' to 'B' since only using blocks of stable helices
 * renaming Predictor*MaxHelix* to Predictor*HelixBlock*
 * IntaRNA/SeedHandler :
   + updateToNextSeed() : generic seed iteration functionality
 * IntaRNA/SeedHandlerIdxOffset :
 * IntaRNA/SeedHandlerExplicit :
 * IntaRNA/SeedHandlerNoBulge :
   + updateToNextSeed() : specialized seed iteration functionality
 * IntaRNA/SeedHandlerIdxOffset :
   + deleteOnDestruction : whether or not the original seedhandler is to be 
     deleted within the destructor
   * destructor() :
     + check for deleteOnDestruction
   * constructor( +deleteOnDestruction = true ) :
     + deleteOnDestruction init

190305 Martin Raden :
 * IntaRNA/PredictorMfe2d :
   * fillHybridE() : 
     + explicit initialization (formerly done via initHybridE)
     * now virtual and to be overwritten by subclasses
   - initHybridE() : obsolete
   - hybridErange : obsolete
 * IntaRNA/PredictorMfe2dSeed : runtime optimization
   + fillHybridE() : overwrites super class function and fills both matrices
     in one go to avoid recomputation of loop energies etc.
   - updateOptima() : obsolete
   - hybridErange : obsolete
   - fillHybridE_seed() : now part of fillHybridE()
 * IntaRNA/PredictorMfe2dHeuristic :
   * predict() :
     - matrix initialization now implicit in fillHybridE()
   * fillHybridE() :
     + explicit base pairing checks and field initialization
 * IntaRNA/PredictorMfe2dHeuristicSeed :
   + fillHybridE() : overwrites super class function and fills both matrices
     in one go to avoid recomputation of loop energies etc.
 * IntaRNA/SeedHandler :
   * isFeasibleSeedBasePair() :
     + check for accessibility constraints
 * IntaRNA/HelixConstraint :
   + checkFullE & evalFullE() : whether or not to use full helix energy for 
     constraint checks or loop terms only
   - useNoED : replaced by checkFullE
 * IntaRNA/HelixHandler* :
   * adaptation to HelixConstraint::evalFullE()
 * renaming PredictorMfe2dHelix* to PredictorMfe2dMaxHelix*, since only maximal
   helices are considered for interaction prediction
 + IntaRNA/HelixHandlerNoBulgeMax : computes for each base pair the canonical 
   helix extension to the right with minimal energy (according the given 
   HelixConstraint) and stores this energy along with the respective helix
   length
 - IntaRNA/HelixHandlerStackingOnly : replaced by HelixHandlerNoBulgeMax
 * IntaRNA/HelixHandler :
   * getHelixHandler() :
     * HelixHandlerStackingOnly replaced by HelixHandlerNoBulgeMax
 + tests/HelixHandlerNoBulgeMax* : renamed HelixHandlerStackingOnly tests
 - tests/HelixHandlerStackingOnly* : renamed
 * bin/CommandLineParsing :
   + helixFullE : replaces helixNoED
   - helixNoED : replaced by helixFullE
   * model : renamed 'H' to 'L' since only using blocks of stable helices
 * README.md :
   + helixFullE documentation 
   * model docu updated

190304 Martin Raden + Frank Gelhausen :
 * API change: E_type now int (was float); all internally used energy values
   are now integer-encoded values in dcal/mol (instead of kcal/mol) as done
   within the VRNA package. this avoids floating point operation issues of the
   traceback routines.
   NOTE: this change will have minor impact on the reported interactions and
   their energies! Thus, results will not equal 2.* predictions!
 + IntaRNA::E_kcal_type = float to represent input/output energies in kcal/mol
 + IntaRNA::Z_type = float to represent partition functions, temperature and
   probabilities
 + type conversion routines for E_kcal_type, E_type, Z_type
 * 'outMaxE' value now exclusive concerning interaction reporting (all reporeted
   interactions have a lower energy)
 * IntaRNA/InteractionEnergyVrna :
   * replacing deprecated 'ViennaRNA::E_Stem(..)' with 'ViennaRNA::vrna_E_ext_stem(..)'
   + Evrna_2_E() to convert VRNA energy values to E_type
 * IntaRNA/NussinovHandler :
   + takes basePairEnergy for minimization (was maximization with hard coded +1 per bp)
 * IntaRNA/PredictorMaxProb :
   + updateOptima(energy) : was missing
 * IntaRNA/PredictorMfe :
   * updateOptima() :
     * now using Interaction::compareEnergy() instead of explicit energy comparison
   * reportOptima() :
     * energy check extended to be conform to Interaction::compareEnergy()
 * IntaRNA/PredictorsMfe2d*Seed :
   * early abortion check corrected
 * IntaRNA/AccessibilityBasePair :
   + VLOG time logging
 * IntaRNA/Accessibility :
   * decomposeByMinPu() :
     + bugfix: explicit minPu==0 handling
 * bin/CommandLineParsing :
   * get*Ranges() :
     + additional Accessibility argument to avoid recomputation (what a mess..)
 * IntaRNA/PredictorMfe2d :
   * initHybridE() : 
     - E_MAX since it was masking global variable with wrong value

##### version 2.4.1

190409 Martin Raden :
 * IntaRNA/Makefile :
   + add missing HelixConstraint.h (thanks to A. R. Garcia and M. Fuchs)

##### version 2.4.0

190228 Martin Raden :
 * easylogging++ :
   * kDefaultLogFileParam now part of header to be accessible
 * bin/IntaRNA :
   * bugfix: check whether or not --default-log-file specified was always true
     such that all log output was going to the default "/dev/null" if not 
     specified by the user 
 * IntaRNA/PredictorMfe2dHelixHeuristic* :
   * bugfix: check for too small interior loops was based on individual gap
     length instead of sum of unpaired positions
 * IntaRNA/PredictorMfe2dHelixHeuristicSeed :
   * updateOptima() :
     + update superclass mfe information to enable early abortion
   * predict() :
     + check if non-seed mfe prediction is within allowed range; otherwise abort
 * IntaRNA/PredictorMfe2dSeed :
   * fillHybridE_seed() :
     - bugfix: wrong assert removed
 * tests updated

190224 Martin Raden :
 * renaming PredictorMfe2dLimStack* to PredictorMfe2dHelix*
 * README.md :
   + general interaction discussion
   + model parameter
   + publication links
   + helix-based prediction + parameters
 + doc/figures/icon-no.*
 + doc/figures/icon-yes.*

190223 Martin Raden / Rick Gelhausen :
 + integration of helix-based RNA-RNA interaction prediction
 + IntaRNA/HelixConstraint : constraints for considered inter-molecular helices
 + IntaRNA/HelixHandler : general interface to enumerate inter-molecular helices
 + IntaRNA/HelixHandlerIdxOffset : allows index shifting for localized access
 + IntaRNA/HelixHandlerStackingOnly : enumerates only canonical helices
 + IntaRNA/HelixHandlerStackingOnlySeed : canonical helices that contain a seed
 + IntaRNA/HelixHandlerUnpaired : enumerate helices that can contain bulges
 + IntaRNA/HelixHandlerUnpairedSeed : (bulged) helices that contain a seed
 + IntaRNA/PredictorMfe2dLimStackHeuristic : helix-based interaction prediction
 + IntaRNA/PredictorMfe2dLimStackHeuristicSeed : helix-based interaction 
   prediction that enforces the seed constraint
 * IntaRNA/SeedHandlerIdxOffset :
   * now subclass of SeedHandler
   + energy access via InteractionEnergyIdxOffset member
 * tests :
   + HelixConstraint
   + HelixHandlerStackingOnly 
   + HelixHandlerIdxOffset + HelixHandlerStackingOnly 
   + HelixHandlerStackingOnlySeed 
   + HelixHandlerIdxOffset + HelixHandlerStackingOnlySeed 
   + HelixHandlerUnpaired
   + HelixHandlerIdxOffset + HelixHandlerUnpaired
   + HelixHandlerUnpairedSeed
   + HelixHandlerIdxOffset + HelixHandlerUnpairedSeed
   + PredictorMfe2dLimStackHeuristic
   + PredictorMfe2dLimStackHeuristicSeed
   + SeedHandlerIdxOffset
   + SeedHandlerMfe
 * bin/CommandLineParsing :
   * replace 'pred' with 'model' = defines underlying interaction model
   + new model 'H' = helix-based prediction
   + new 'Helix' parameter group

190207 Martin Raden :
 * IntaRNA/OutputHandlerText :
   * detailed output reordered and extended
     * Pu now below respective ED
     * Pu now with ':' prefix instead of '+' (thanks to Sebastian Will)
     + E(hybrid) information (thanks to Sebastian Will)
   * NOTE: this causes changes in the detailed ASCII output!
 * README.md :
   * minor fixes
 * IntaRNA/OutputHandlerText :
   * add() :
     * explicit handling of duplicated base pairs (might happen due to explicit
       seeds containing only a single base pair) (thanks to Sebastian Will)

190128 Martin Raden :
 * bump easylogging++ version to v9.96.7

190110 Martin Raden :
 * CommandLineParsing :
   * windowOverlap : default changed to 150 (was 0) to avoid the need to set for
     default q|tAccW values

##### version 2.3.1

181108 Martin Raden :
 + IntaRNA/PredictionTrackerSpotProbAll : exhaustively computes and prints all
   spot probabilities (ie for all intermolecular index pairs)
 * bin/CommandLineParsing :
   + registration and documentation of PredictionTrackerSpotProbAll usage
 * README.md :
   * exhaustive spot prob calculation documented

180907 Martin Raden :
 * IntaRNA/OutputHandlerCsv :
   + 'hybridDP|Bfull' output for full sequence lengths
 * IntaRNA/Interaction :
   * dotBar() :
   * dotBracket() :
     + "fullLength" argument to trigger full-length or interaction-site-only 
       output
 * README.md : 
   + 'hybridD(P|B)full' documentation

##### version 2.3.0

180906 Martin Raden :
 * bin/CommandLineParsing :
   * VRNA version information in --version output corrected
 * README.md :
   + window-based prediction docu
   * minor corrections and extensions

180905 Martin Raden :
 * configure.ac :
   + final configuration output
 * IntaRNA/PredictorMfe4d : 
   * predict() :
     * bugfix initialization problem : was only initializing cells not to be 
       filled, which caused random prediction problems (issue #117)
     - debug cell allocation information
     - sophisticated (but buggy) check for cells not needed for computation
   * fillHybridE() :
     * bugfix : maxInteractionWidth check (have to use <= instead of <)
     - cell initialization check removed (since obsolete and causing problems,
       see issue #117)
 * IntaRNA/PredictorMfe4dSeed : 
   * predict() :
     * bugfix initialization problem : was only initializing cells not to be 
       filled, which caused random prediction problems (issue #117)
     - debug cell allocation information
     - sophisticated (but buggy) check for cells not needed for computation
 * README.md :
   * prediction mode documentation corrected
   
180905 Martin Raden, Sebastian Holler : window-based prediction enabled
 * bin/CommandLineParsing :
   + windowWidth parameter + validator to define length of windows
   + windowOverlap parameter + validator to define overlap of subsequent windows
   * parse() :
     + checks for window* parameter dependencies/interactions
 * bin/IntaRNA :
   + window-decomposition of ranges
   + parallelization of window-based processing if only one query/target
 * IntaRNA/IndexRange :
   + overlappingWindows() : provides a vector of overlapping windows for 
     window-based processing
 * IntaRNA/OutputHandlerInteractionlist :
   * add() : 
     + check for identity before insert to avoid duplications in list
 * tests/IndexRange :
   + check for overlappingWindows()
 + tests/OutputHandlerInteractionList :
   + order test
   + duplicated insert test

180819 Martin Raden :
 + IntaRNA/SeedHandlerNoBulge : optimized seed handler if no bulges are allowed
   within seeds
 + tests/SeedHandlerNoBulge_test : comparing output with SeedHandlerMfe
 * IntaRNA/SeedHandler :
   + isFeasibleSeedBasePair() : checks whether or not a given base pair can be
     part of a seed (complementary, in valid region, with ED below threshold)
 * bin/CommandLineParsing :
   * getSeedHandler() :
     + provides SeedHandlerNoBulge if no unpaired bases are allowed within seed
 * IntaRNA/SeedHandlerMfe :
   * fillSeed() :
     + time measurement in verbose mode
     + usage of SeedHandler::isFeasibleSeedBasePair() where appropriate

180807 Martin Raden :
 * Makefile.am :
   + set GCC_COLORS (if not already present) to enable gcc output coloring

180711 Martin Raden :
 * IntaRNA/Interaction :
   * operator < : 
     * bugfix: right base pair access was incorrect (thanks to Sebastian Holler)

180710 Martin Raden :
 * IntaRNA/Predictor* :
   * traceback() :
     * sanity check corrected (debug mode only) to handle single bp interaction

180705 Martin Raden :
 + IntaRNA/PredictionTrackerProfileSpotProb : computes a profile for each 
   sequence position about the probability that a position is covered by an
   interaction (but not necessarily base paired within) = spot probability
 * IntaRNA/Makefile :
   + PredictionTrackerProfileSpotProb
 * bin/CommandLineParsing :
   + PredictionTrackerProfileSpotProb
   * '--out' argument :
     + q|tSpotProb : generate spot probability profile for query|target
   * getCodeForOutPrefix() :
     + parsing of "q|tSpotProb"
 * README.md : 
   + docu of spot probability profiles 
 
##### version 2.2.1

180627 Martin Raden / Milad Miladi :
 * adaptation to VRNA v2.4.8
    
180328 Martin Raden :
 * IntaRNA/OutputHandlerCsv :
 * IntaRNA/OutputHandlerIntaRNA1 :
 * IntaRNA/OutputHandlerText :
     + data first written to local stream before pushed (via OMP critical) to
       final output stream. This should fix some multi-threading output issues.
 * IntaRNA/OutputHandlerText :
   + usage of reportedInteractions member
   * destructor() :
     + check if nothing was reported and provide according output
 * bin/CommandLineParsing :
   * "version" output extended with used VRNA and boost version
     
##### version 2.2.0

180319 Martin Raden :
 * IntaRNA/PredictorMfe2dHeuristic :
   * traceBack() :
     * bugfix: wrong sanity check in debug mode (thanks to Rick Gelhausen)
 * IntaRNA/Predictor :
   * updateOptima() :
     * bugfix: PredictionTracker call was not doing lastPos-handling
 * README.md : 
   + spotProb docu added
 + IntaRNA/PredictionTrackingSpotProb : tracks for a given set of interaction
   spots their probability to be covered/spanned by an interaction. furthermore,
   the probability that none of the spots of interest is covered is provided.
 + test/PredictionTrackingSpotProb
 * bin/CommandLineParsing :
   * OutputPrefixCodes :
     + OP_spotProb : for spotProb output mode
   + outSpotProbSpots : member to store the spots to be tracked (if provided)
   * --out argument :
   * validate_out() :
   * getPredictor() : 
     + handling of "spotProb" prefix for spot probability tracking

180312 Martin Raden :
 * IntaRNA/PredictorMaxProb :
 * IntaRNA/PredictorMfe :
   * updateOptima() :
     * bugfix : call of predTracker->updateOptimumCalled() was done without
       index offset information (fixes github issue #110) 
 * IntaRNA/AccessibilityVrna :
   * callbackForStorage() :
     * bugfix : copying was not heeding maximal interaction length 
       (fixes github issue #112)
 * configure :
   * bump VRNA dependency to v2.4.4 for recent SHAPE-related bugfix
 * README.md :
   + SHAPE-reactivity-related section 
 * IntaRNA/AccessibilityConstraint :
   + regexShapeMethod : regex for shapeMethod parsing
   + regexShapeConversion : regex for shapeConversion parsing
   + shapeMethod member + getter
   + shapeConversion member + getter
   * constructor :
     + argument for shapeMethod/shapeConversion
     + sanity check of shapeMethod/shapeConversion if shapeFile non-empty
   * copyconstructor :
     + copying of shape* members
   * regex : bugfix : was not correctly encoding dotbracket strings or index
     range lists
   * operator= :
     + shape members copied
 * tests/AccessibilityConstraint :
   * adaptation to AccessibilityConstraint constructor changes
   + tests for regex matching
 * IntaRNA/Accessibility :
   * adaptation to AccessibilityConstraint constructor changes
 * IntaRNA/AccessibilityVrna :
   * fillByRNAplfold() :
     + get shapeMethod and shapeConversion from AccessibilityConstraint
 * bin/CommandLineParsing :
   * adaptation to AccessibilityConstraint constructor changes
   + SHAPE argument list
     * q|tShape : moved here from query|target argument list
     + q|tShapeMethod : method to convert SHAPE to pseudo energies
     + q|tShapeConversion : method to convert SHAPE to pairing probabilitie
   * validate_structureConstraintArgument() :
     * bugfix: check had to be reversed

1802 Martin Raden :
 * IntaRNA/AccessibilityConstraint :
   * constructor() :
     - default arguments removed to avoid wrong constructor selection
     + shapeFile argument to pass file name of SHAPE reactivity data to be used
   + shapeFile member + getter
   * isEmpty() :
     + check if shapeFile empty
   * operator<< :
     + output of shapeFile if non-empty
 * IntaRNA/Accessibility :
   * adaptation to AccessibilityConstraint constructor changes
 * IntaRNA/AccessibilityVrna :
   * fillByRNAplfold() :
     + call vrna_constraints_add_SHAPE() if shapeFile of AccessibilityConstraint
       not empty
 * bin/CommandLineParsing :
   + q/tShape argument + validation : file name from which to read SHAPE
     reactivity data to guide accessibility computation (via soft constraints)
   * adaptation to AccessibilityConstraint constructor changes
   * unification of parsing code update via usage of updateParsingCode()

180111 Martin Raden :
 * IntaRNA/Accessibility:
   + decomposeByMinPu() : decomposes a given list of ranges to subranges by
     excluding all positions that have an unpaired probability below a given
     threshold
 * bin/CommandLineParsing :
   + '--outMinPu' argument + validators
   * get*Ranges() :
     + calling Accessibility->decomposeByMinPu() for each range list
     + energy argument (needed for RT access)

171216 Martin Raden :
 * bin/CommandLineParsing :
   - check for equal length of sequence and constraint removed since obsolete

171214 Martin Raden :
 * m4:
   * *.m4 : updated to recent versions
   + m4-ax_boost_*.m4 : boost library checks 

171213 Martin Raden :
 * configure.ac :
   * leading spaces in macros removed
   + default macro definitions if pkg.m4 not available
   + silent make output enabled per default (full output with 'make V=1')
   - obsolete pkg-config m4 check removed

171211 Martin Raden :
 * IntaRNA/PredictorMfe :
   * traceBack() : 
     + argument outConstraint (needed to bugfix the PredictorMfe2d computation)
   * reportOptima() :
     + forwards outConstraint to traceBack() call
 * IntaRNA/PredictorMfe* :
   * adaption to traceBack() interface changes
 * IntaRNA/PredictorMfe2d :
   * traceBack() :
     * bugfix: outConstraint missing in fillHybridE() call but called with 
       (&outConstraint = i1) and (i1 = i2), which caused infinite tracebacks
   * initHybridE() :
     - default argument values (=0) removed to avoid wrong argument use (see
       traceBack() bugfix)
     - special initialization removed to simplify code (and not applied anyway)
   * fillHybridE() :
     - default argument values (=0) removed to avoid wrong argument use (see
       traceBack() bugfix)
     * computation simplified
 * IntaRNA/PredictorMfe2d :
   * fillHybridE() :
     * bugfix: outConstraint missing in fillHybridE() call but called with 
       (&outConstraint = i1) and (i1 = i2), which caused infinite tracebacks
 * IntaRNA/PredictorMfe2dSeed :
   * fillHybridE_seed() :
     + outConstraint argument
     * bugfix: outConstraint missing in fillHybridE() call but called with 
       (&outConstraint = i1) and (i1 = i2), which caused infinite tracebacks

171127 Martin Raden :
 * bin/CommandLineParsing :
   * getThreads() :
     * returns threads.max if threads.val == 0
   * threads :
     * min value = 0 (request maximal number of threads)
     * parameter docu updated accordingly
   * validateRegion() :
     + instantiate range list to ensure that it is ascending and non-overlapping
   + outPerRegion : parameter flag to define how reporting per region should be
     done
   + reportBestPerRegion() : access to outPerRegion
 * IntaRNA/IndexRangeList :
   * sanity checks now also in non-debug mode to ensure sane instances
 * IntaRNA/IndexRange :
   * operator+/-(int shift) : returns a shifted range
 * bin/IntaRNA :
   * bugfix: verbose range report was not shifting ranges by +1
   + use OutputHandlerInteractionList to handle predictions for all region
     combinations for a query-target combination; setup maximal size based on 
     CommandLineParsing::reportBestPerRegion()
 * README.md :
   + '--outPerRegion' docu added
 * IntaRNA/Interaction :
   + operator < : less-than comparison based on energy and indexing/#bp/seed.E
 + IntaRNA/OutputHandlerInteractionList : stores the best k interactions sorted
   by Interaction::operator<()

171123 Martin Raden :
 * bin/CommandLineParsing :
   * validate_outputTarget() :
     + check whether or not parent directory of output file exists
     + check whether or not the output file can be overwritten (if existing)
 * IntaRNA/general :
   * E_precisionEpsilon : now = sqrt(numeric_limits<E_type>::min())
     (was 1000*numeric_limits<E_type>::epsilon(), which was much too large)
 * IntaRNA/AccessibilityVrna :
   * callbackForStorage() :
     * undo last change, i.e. check for 0-pUvalues now again via '==' instead of
       E_equal
 * IntaRNA/AccessibilityVrna :
 * IntaRNA/InteractionEnergyVrna :
   * construction() :
     * now computation is considered thread-safe, i.e. no omp_critical needed
 * bin/IntaRNA :
   + parallelization of query accessibility computation
 * README.md :
   - non-parallelization of accessibility computation not true anymore

171122 Martin Raden :
 * README.md :
   + discussion and examples for range-based constraint encoding
 * tests/AccessibilityConstraint :
   + test for range-based encoding parsing
 * IntaRNA/IndexRangeList :
   + regexString : string encoding of regex without start/end matching to be
     used in other regex encodings
 * IntaRNA/AccessibilityConstraint :
   + dotBracket_constraints : list of constraint encodings
   + regionIndexList : regular expression string that matches the region list
     for a single constraint encoding
   + regex : overall regex that matches both dotBracketAlphabet as well as
     range-based encodings
   + operator << : prints the constraint in range-based encoding
   * construction :
     + length now mandatory since not necessarily available via constraint 
       string encoding length
     + parsing of range-based encoding with according error messages
 * bin/CommandLineParsing :
   * get*Accessibility():
     + constraint parsing now encapsulated by try-catch to report dedicated
       error messages for query/target encodings
   + *AccConstr parameter description extended with range-based encoding
   * validate_structureConstraintArgument() :
     - obsolete structure check removed
     * validation now via regex

171117 Martin Raden :
 * bin/CommandLineParsing :
   * unify input error handling via throw
   + check if accessibility available for RegionLenMax-based decomposition
   * validate_*AccConstr() :
     - check that not in plfold-based mode
   * AccConstr :
     + paired constraint option
 * IntaRNA/VrnaHandler : 
   + modle.compute_bpp = 1 to enable correct ED computation via plfold routines
 * IntaRNA/AccessibilityVrna :
   * construction() : 
     - checks to enable constraints only for RNAup-based computation
     - RNAup-based computation option removed
     * plfold-based computation only
   * callbackForStorage() :
     * prob_unpaired == 0 check via E_equal()
   * fillByRNAplfold() :
     - debug check that no constraint is defined
     + constraint setup
 * README.md :
   + --q/tAccConstr description
 * IntaRNA/AccessibilityConstraint :
   + dotBracket_unconstrained : holds unconstrained encoding char
   + dotBracket_paired : holds paired encoding char
   * dotBracket_alphabet : now compiled from static vars only (excludes round
     brackets that were part of it so far but never enabled)
   - is*(from,to) : obsolete (and wrong) checks for regions removed
   + paired : list of ranges marked as paired
   + isMarkedPaired(i) : whether or not a position is marked as paired
   * constructor() :
     + parse paired ranges
     - check for round brackets (obsolete)
   * is*() : take paired into account
   * getVrnaDotBracket() :
     + returns '|' for paired positions
 * IntaRNA/Accessibility* :
   * getED() :
     * bugfix : do accessibility check only on interval ends (not whole region)
 * tests/AccessibilityConstraint_test :
   + test for position marked as paired

171116 Martin Raden :
 * configure.ac :
   + --enable-static-linking flag added to build statically linked binaries
 * IntaRNA/PredictorMfe2dHeuristic :
   * undo change of 171111 since it is effecting seed-based predictor as well
     and it is not benchmarked yet.. (thanks to Alex Gawronski)

171113 Martin Raden :
 * IntaRNA/IndexRangeList :
   + get() : (constant) access to a certain range
 + tests/testEasyLoggingSetup.icc : easylogging setup to be included in all
   tests to enable (debug) logging from the IntaRNA lib
 * tests/*_test.cpp :
   + inclusion of testEasyLoggingSetup.icc
 * Makefile.am :
   + clean test executable to ensure rebuild
 * tests/Makefile.am :
   + easylogging++ compiler flags
 * IntaRNA/Accessibility :
   + decomposeByMaxEd() : automatically identifies highly accessible regions of
     the sequence and provides list of according intervals where each is below
     a given threshold in length (also lower bound provided). This is done by
     iteratively identifying the max ED window (of given size) within too long
     ranges and removing it, which results either in a shrink of the range or
     a decomposition into two ranges. This is repeated until all ranges are 
     sufficiently short.
     (thanks to Alex Gawronski)
 * tests/AccessibilityFromStream_test :
   + test Accessibility.decomposeByMaxEd()
 * bin/CommandLineParsing :
   + qRegionLenMax/tRegionLenMax : (optional) maximal length of automatically
     identified highly accessible regions in query/target
   * qRegion/tRegion : mutable to enable lazy computation in get*Region()
   * get*Region() :
     + lazy computation of regions if according ?RegionLenMax > 0; uses a 
       winSize == seedBP and a minimal range length of seedBP
   * parse() :
     + ensure that no manual regions defined if ?ReginLenMax provided 

171111 Martin Raden :
 * IntaRNA/PredictorMfe2dHeuristic :
   * init curCellEtotal = E_INF to ignore effects of low ED of single positions
     (already applied that way in PredictorMfe2dHeuristicSeed)

171110 Martin Raden :
 * CommandLineParsing :
   + qSet : parameter to define sequence subset to be considered for prediction
   + tSet : parameter to define sequence subset to be considered for prediction
     (thanks to Alex Gawronski)
 * IndexRange :
   + LAST_INDEX : placeholder for last available index
 * README.md :
   + --qSet/--tSet docu added
   * fixing easylogging lib use example

171019 Martin Raden :
 * configure.ac :
   + in debug mode, set ELPP_DEBUG_ERRORS compiler define
 * IntaRNA/general.h :
   - ELPP_* defines removed since set via configure or not needed
 * bin/CommandLineParsing :
   + argument "--default-log-file" to redirect log output to file (except ERROR)
   + logFileName : dummy variable for --default-log-file argument
 * README.md :
   + --default-log-file docu added
   + -v/--verbose docu added
 * bin/IntaRNA :
   + log-file handling
   + flushing log streams on exit

##### version 2.1.0

171010 Martin Raden :
 * README.md :
   + explicit seed description
 * AccessibilityVrna :
   * fillByConstraints() :
   * fillByRNAup() :
   * callbackForStorage() :
     * BUGFIX : only check for block-constraint for region ends (is allowed to
       contain a blocked subregion)

171009 Martin Raden :
 * tests/*BasePair_test :
   * explicit use of std::abs (should fix test error of github issue #91)
   * BUGFIX: test values wrong (were derived from rounded values)
     -> fixes github issue #91 (thanks to @sjannielefevre) 
 * AccessibilityBasePair :
   * constructor() :
     + RT sanity check to avoid division by zero
 * NussinovHandler :
   + printMatrix() : utility function for debug to see matrix content
   * getQ(), getQb(), getPbp(), getPu() :
     * BUGFIX : boundary checks partially wrong
     + explicit lower bound handling
 + SeedHandlerMfe : former SeedHandler implementation for mfe-seed access
 * SeedHandler : now abstract interface
 * Predictor*Seed :
   * constructor() :
     * takes now directly a SeedHandler instance instead of a SeedConstraint
 * CommandLineParsing :
   + getSeedHandler() : provides the seed handler according to the user input
   * getPredictor() :
     * uses now getSeedHandler() to setup Predictor*Seed instances
   + '--seedTQ' argument for explicit seed encodings
 * SeedHandlerMfe() :
   * constructor() : sanity check in debug mode 
 + SeedHandlerExplicit() : seed handler for explicitly defined seed interactions
   + checkSeedEncoding() : tests sanity of seed encoding
   + SeedData() : data structure that parses and holds explicit seed information
 * SeedConstraint :
   + explicitSeeds + getter : explicit seed encodings to be used
 + tests/SeedHandlerExplicit :
   + test checkSeedEncoding()
   + test SeedData() parsing

171006 Martin Raden :
 * AccessibilityVrna :
   * fillByRNAup() :
     * BUGFIX : have to check whether region is blocked not if constrained
 * AccessibilityBasePair :
 * AccessibilityFromStream :
   + "not implemented" notification if accessibility constraints given

171005 Martin Raden :
 * configure.ac :
   + disable-log-coloring : sets INTARNA_LOG_COLORING for console log coloring
 * bin/IntaRNA :
   * colored log output depends now on INTARNA_LOG_COLORING macro
 * IntaRNA/AccessibilityVrna :
   + callbackForStorage() : callback function to use new VRNA API
   * fillByRNAplfold() : window-based PU computation via new VRNA API

171004 Martin Raden :
 * configure.ac :
   * bump minimal ViennaRNA version to 2.4.1 (for constraint interface) 
   * bump easylogging++ version to 9.95
   + easylogging preprocessor macros to get new version compiled
 + src/easylogging++.cc : source code of new easylogging++ version
 + src/easylogging++.LICENCE.txt : licence information of easylogging++
 + lib 'easylogging' since new version comes with source file
 * tests : adaptation to new easylogging++ version and library
 * bin/IntaRNA : 
   * adaptation to new easylogging++ version and library

##### version 2.0.5

170926 Martin Raden :
 * IntaRNA/PredictorMfe4d :
   * getNextBest() : BUGFIX : matrix access was not using interaction end 
     distance and interaction data was wrongly storing distance as right index;
     this was causing segfaults when requesting more than one prediction output
     (thanks to Rick Gelhausen)
 
170926 Martin Raden :
 * IntaRNA/PredictorMfe4dSeed :
 * IntaRNA/PredictorMfe2dHeuristicSeed :
   + updateOptima() to disable predTracker updates when fillHybridE() of super
     class is called; 
     --> BUGFIX : so far resulting tracker profiles were a mix of non-seeded and
         seeded interactions and thus not complementary with the predicted
         interactions

##### version 2.0.4

170706 Martin Raden :
 * IntaRNA/OutputHandlerCsv :
   + E_hybrid : hybrid-only energy = E - ED1 - ED2
   + E_norm : length-normalized energy = E / ln( length1*length2 )
   + E_hybridNorm : length-normalized hybrid-only energy =  E_hybrid / ln(l1*l2)

170705 Martin Raden :
 * perl/Intarna*.pl :
   * shebang corrected (thanks to Patrick Wright)

170503 Martin Mann :
 * README.md :
   * output file description updated
   + description of easylogging++ initialization for library usage 
     (thanks to Maria Waldl) 
 * bin/CommandLineParsing :
   * getFullFilename() : 
     * bugfix : handling of paths in file name definitions
       (thanks to Milad Miladi)
     * sequence index now suffix of filename (added in front of file extension) 

##### version 2.0.3

170503 Martin Mann :
 * README.md :
   + OS X install instructions (thanks to Lars Barquist)
 + IntaRNA/NussinovHandler : (thanks to Mostafa Mahmoud) implementation of
   intramolecular structure prediction, partition function and probability
   computation using a variant of the Nussinov-Jacobson model
 * IntaRNA/AccessibilityBasePair :
   * uses NussinovHandler for computation
 * IntaRNA/InteractionEnergyBasePair :
   * uses NussinovHandler for computation
   + computeES() : fill a given ES value container
   * getES*() : enabled
 + tests for NussinovHandler 
 + tests for InteractionEnergyBasePair
 + doc/latex-deps : provides required doxygen-latex dependencies

170413 Martin Mann :
 * doc/* 
   + added to distribution build

170412 Martin Mann :
 * configure.ac :
   * doxygen docu generation now truely optional
   * pkgconfig usage now truely optional (check for m4 functions)
 * Makefile.am :
   * doxygen docu generation now truely optional
   - include of doc/doxygen.am obsolete
   + @DX_RULES@ : doxygen rules generated by configure
 - doc/doxygen.am : obsolete

170411 Martin Mann :
 * PredictorHeuristicSeed :
   * predict() :
     * bugfix : reinit was not overwriting all hybridE_seed entries 
       (thanks to Alex Gawronski)

##### version 2.0.2

170328 Martin Mann :
 * PredictorHeuristic* :
   * predict() :
     * bugfix : reinit was not overwriting all matrix entries 
       (thanks to Alex Gawronski)

170307 Martin Mann :
 * header defines prefixed with INTARNA_
 * general.h :
   + checks if defines with general names are already present
   * intarna specific defines prefixed with INTARNA_
 * configure.ac :
   * VRNA configure flag conform with LocARNA (--with-vrna)
 * README.md
   + (minimal) library information added
 * CommandLineParsing :
   + argument boundaries for outDeltaE and outMaxE
 * SeedHandler :
   * traceBackSeed() : 
     * bugfix: was not using offset1/2

170306 Martin Mann :
 + src/IntaRNA : subfolder that holds all libIntaRNA related sources
 + "IntaRNA" namespace for all libIntaRNA sources
 + libIntaRNA.a build
 * tests adapted to library usage
 * configure.ac :
   + pkg-config based VRNA version check (optional)
   + --disable-pkg-config switch
   * tests rewritten using AS_IF()
   + AC_CONFIG_MACRO_DIR() replaces individual m4_include() calls
   + VRNA_REQUIRED_VERSION variable
   + AC_PROG_RANLIB
   + pkg-config related files and commands

170227 Martin Mann :
 * InteractionEnergy :
   + getE_multi() : generic function to get the energy contribution of a 
     intramolecularly structured gap between two interaction sites
 * InteractionEnergy :
 * InteractionEnergyBasePair :
 * InteractionEnergyIdxOffset :
 * InteractionEnergyVrna :
   - getEU() : renamed to getE_multiUnpaired()
   + getE_multiUnpaired() : former getEU()
   + getE_multiHelix() : penalty for the intermolecular site that is part of the
     intermolecular multiloop as a helix (right interaction site)
   + getE_multiClosing() : penalty for closing the multi-site gap (on the left)

##### version 2.0.1 

170221 Martin Mann :
 * PredictionTrackerPairMinE :
   * writeData() :
     * column order reversed to be index ascending

170219 Martin Mann :
 * RnaSequence :
   + operator == : equality check (id,length,seq)
 * PredictionTrackerPairMinE :
   * writeData() :
     + col/row header now includes sequence position beside according nucleotide
 * PredictionTrackerProfileMinE :
   * writeProfile() :
     + additional column for according nucleotide
 * CommandLineParsing : 
   + getFullFilename() : provides the fill in/output file name given a suffix,
     i.e. adding a sequence specific prefix if needed
   * validate_?AccFile() :
     + getFullFilename() usage for multi-sequence accessibility input
     + additional check to hinder STDIN-based multi-sequence accessibility input
   * write*Accessibility() :
     + getFullFilename() usage for multi-sequence accessibility output
   * construction() :
     - ?AccFile : no validation on parsing (requires prior full sequence parse)
   * parse() :
     + ?AccFile validation after sequence parsing is done
     * bugfix: *AccConstr presence check corrected
     * *Acc : accessibility load from file for multi-sequence input enabled
     - generation of additional output files for multi-sequence input enabled
   * get*Accessibility() :
     + getFullFilename() usage for multi-sequence accessibility input
   * getPredictor() :
     + getFullFilename() usage for multi-sequence prediction tracking output
   

170217 Martin Mann :
 * README.md :
   + outfile section
   + minE profile docu
   * accessibility output corrected
 + doc/figures/profile-minE.png : minE profile example
 * CommandLineParsing :
   - out*File : replaced by --out multi-arg handling
   - validate_out*File() : obsolete
   + OutPrefixCode : encodings for all valid prefixes of --out
   + getCodeForPrefix() : provides the OutPrefixCode for a given prefix string
   + outPrefix2streamName : holds for each registered --out prefix the user
     provided file/stream name or an empty string if not provided
     (filled by validate_out())
   * out : now a vector of arguments to enable multi-arg occurrences
   * write*Accessibility : uses now outPrefix2streamName values
   * constructor :
     + initialization of outPrefix2streamName
     * argument help unified
   * parse() :
     + duplication check for --out arguments to avoid file write clashes
   + PredictionTrackerPairMinE enabled
 * Interaction :
   + Seed : class to concentrate all seed-related information
   - seedRange : replaced by seed member
   + seed : optional Seed object
   * setSeedRange() : fills seed member
   + operator=(Interaction) : copy function that creates new seed object
   + copy constructor: to copy seed object
 * OutputHandler* :
   * adaption to Interaction::Seed change

170214 Martin Mann :
 + PredictionTrackerPairMinE : tracks for all intermolecular index pairs the
   minimal energy of any interaction covering this pair
 + PredictionTrackerHub : forwards tracking information to all registered
   trackers
 * CommandLineParsing : 
   * getPredictor() :
     + PredictionTrackerHub used for tracker setup

170212 Martin Mann :
 * general :
   + newOutputStream() : returns a pointer to a (newly allocated) output 
     stream depending on the provided string
   + deleteOutputStream() : deletes (if necessary) the pointer returned by
     newOutputStream
 * CommandLineParsing :
   * bugfix : -n argument description wrong [thanks to Michael Uhl]
   + out*minEFile + validator : stream name of minE profile output
   * destruction :
     + using deleteOutputStream()
   * parse() :
     + using newOutputStream()
     + checks for out*minEfile
   * getPredictor() :
     + PredictionTracker setup
   * writeAccessibility() :
     + using newOutputStream()
     + using deleteOutputStream()
 * InteractionEnergy :
   + getE( Z ) : ensemble energy from partition function
 + OutputHandlerHub : hub for a list of OutputHandler objects to that the output
   requests are forwarded
 + PredictionTracker : general interface to track all step results produced by
   a Predictor instance
   + updateOptimumCalled() : triggered when a Predictor calls updateOptimum()
 + PredictionTrackerProfileMinE : produces a profile of the minimum interaction
   energy for each position of the interacting RNAs (covered by any interaction)
   + stream-based construction
   + filename/streamname-based construction
 * Predictor :
   + predTracker : optional PredictionTracker pointer
   * constructor :
     + predTracker
   * destruction :
     + delete predTracker
 * PredictorMfe :
   * updateOptimum() :
     + calling predTracker->updateOptimumCalled() if non-NULL predTracker
 * PredictorMfe* :
   * constructor :
     + predTracker
 * PredictorMaxProb* :
   * constructor :
     + predTracker
   * updateOptimum() :
     + calling predTracker->updateOptimumCalled() if non-NULL predTracker
   * using energy.getE( Z ) 
 + tests : PredictionTrackerProfileMinE
 * tests : rename OutputRangeOnly -> OutputHandlerRangeOnly

##### version 2.0.0 

170206 Martin Mann :
 * src/Makefile.am :
   + AUTOMAKE_OPTIONS = std-options : checks for std options of binary
 * tests/Makefile.am :
   * tests now build via 'check' make target (autotools-based setup)
 * Makefile.am :
   * adaptions to test Makefile changes

170203 Martin Mann :
 * configure.ac : 
   + check for VRNA >= 2.3.0 via function 'vrna_md_copy' exists
 * AccessibilityConstraint
   + maxBpSpan + getter : maximal base pair span to be considered for 
     accessibility computation
 * Accessibility* :
   - getES() : now part of InteractionEnergy interface
 * AccessibilityVrna :
   - EsMatrix : moved to InteractionEnergyVrna
   - esValues : moved to InteractionEnergyVrna
   - computeES() : moved to InteractionEnergyVrna
   * constructor() :
     - plFoldL : now via AccessibilityConstraint.getMaxBpSpan()
     - computeES
   * ED >= 0 ensured
 * InteractionEnergy :
   * getES*() now abstract and to be implemented in subclass
   - getES*() implementation based on Accessibility interface
 * InteractionEnergyBasePair :
   + getES*() : throws exception since not implemented yet
   * constructor() :
     + initES : whether or not to compute precompute ES values
 * InteractionEnergyVrna :
   * constructor() :
     + initES : whether or not to compute precompute ES values
   + EsMatrix data type
   + esValues1/2 : optional ES value data
   + getES*() : returns esValues* entry if available; throws exception otherwise
   + computeES() : inits ES values based on VRNA routines
 * VrnaHandler :
   * getModel() :
     * uses now vrna_md_copy() (VRNA >= 2.3.0) for model copying
 * CommandLineParsing :
   * energy : 
     - 'F' : replaced by 'V'
     + 'V' : VRNA based computation
   - OutputMode : replaced by char encoding
   * outMode : now char encoding
 * OutputHandlerText :
   + detailedOutput : flag to switch between detailed and reduced output
   * add() : 
     + heeds detailedOutput flag
   * construction() :
     + detailedOutput flag handling
 * PredictorMfe2d :
   * initHybridE() :
     + outConstraint argument
     * bugfix : maxE check was not considering outConstraint.maxE
   * fillHybridE() :
     + outConstraint argument
 * PredictorMfe4d* :
   * predict :
     * bugfix : maxE check was not considering outConstraint.maxE

170202 Martin Mann :
 * CommandLineParsing :
   * outMaxE maximum now 999 (was 0)
   - PredictionMode : replaced by char-based setup
   * seedMaxUPq -> seedQMaxUP
   * seedMaxUPt -> seedTMaxUP
   * seedRangeq -> seedQRange
   * seedRanget -> seedTRange
   + pred + validator : setup of prediction target (mfe-si, maxprob-si, ...)
   * predMode : 
     * char-based setup
     - max-prob mode (now selected via --pred=P)
   * outOverlap : char-based setup
   * outAccFileq -> outQAccFile
   * outAccFilet -> outTAccFile
   * outPuFileq -> outQPuFile
   * outPuFilet -> outTPuFile
   * get*Accessibility() :
     * computeES setup now via pred.val
   * getPredictor() :
     + support for different prediction targets via pred.val switch
 * OutputConstraint :
   * constructor() :
     - (maxE < 0) check removed
 + intarna_config.h.in : header containing package version and configure 
   information
 * configure.ac : 
   + generate intarna_config.h
   + AC_SUBST(INTARNA_MULITHREADING)
 * AccessibilityDisabled :
   * getES() : 
     - no error is thrown anymore
 * perl/*.pl : adaptions to CLI changes

170201 Martin Mann :
 * CommandLineParsing :
   * max sequence number now 99999
   * setup of threads parameter only if INTARNA_MULITHREADING
 * configure.ac : 
   + --disabled-multithreading : sets INTARNA_MULITHREADING to 0, otherwise 1
 * enclose all OMP stuff with INTARNA_MULITHREADING preprocessor checks
 * AccessibilityVrna :
   * computeES() : bugfix
 
170131 Martin Mann :
 * CommandLineParsing :
   + opts_cmdline_short : short option list for abbreviated help output
   * constructor() :
     + setup of opts_cmdline_short
   * parse() 
     + handling of short and full help output
     + disallow multi-threading for IntaRNA v1 output (due to separators)
   * q|tAcc : renamed 'F' option to 'C' for 'computation of accessibilities'
   * adaption to OutputHandlerIntaRNA1* rename
   + OutputMode::V1_NORMAL : normal version 1.* output
   + threads + validate + getter : OMP max thread number
 - OutputHandlerIntaRNA1detailed : renamed to OutputHandlerIntaRNA1
 + OutputHandlerIntaRNA1 : former OutputHandlerIntaRNA1detailed
   + detailedOutput : flag to switch between detailed and normal v1 output
   + OMP : ensure no parallel write to output stream
 * perl/IntaRNA*_1ui.pl :
   + normal output enabled (-o flag now with effect)
 * intaRNA :
   * adaption to OutputHandlerIntaRNA1* rename
   + OMP : parallelization of target/query processing
   + OMP : exception handling within parallelized loops due to missing OMP 
     exception forwarding
   * final exception handling now always returns -1 (even in debug mode)
 * OutputHandlerCsv : 
   + E_loops output
   + OMP : ensure no parallel write to output stream
 * OutputHandlerText : 
   + OMP : ensure no parallel write to output stream
 * AccessibilityVrna :
   + OMP : ensure no parallel VRNA-calls since not threadsafe
 + m4/m4_ax_openmp.mp4 : OMP checks
 * configure.ac :
   - backward compatibility stuff for autoconf < 2.65 
   + OMP check and AM_CXXFLAGS setup

170130 Martin Mann :
 * CommandLineParsing :
   + stdinUsed : flag to ensure STDIN is used for input only once
   + setStdinUsed() : sets stdinUsed = true if not already or throws log 
     otherwise
   * q/tAcc : 
     + 'P' : unpaired probabilities in RNAplfold style from --q/tAccFile
     + 'E' : ED values in RNAplfold-Pu-like style from --q/tAccFile
   + validate_q/tAccFile() 
   * get*Accessibility() :
     + AccessibilityFromStream support
     * bugfix : maxLength setup has to be = min(maxLength,accW)
   + outAccFile* + validate : output ED values to file/stream
   + outPuFile* + validate : output Pu values to file/stream
   + validate_outputTarget() : generic text function for stream names
   * validate_out() : calls validate_outputTarget()
   + write*Accessibility() : write the given accessibility to file/stream if
     requested by the user
   + writeAccessibility() : writes accessibility data to file/stream in the 
     requested format
   * q/tIntLenMax argument info :
     + information that --q/tAccW might overwrite
 * Accessibility :
   * getMaxLength() : return value non-const size_t
   + writeRNAplfold_ED_text() : Pu output in RNAplfold style
   + writeRNAplfold_Pu_text() : ED output in RNAplfold style
   + writeRNAplfold_text() : generic RNAplfold Pu-styled text output
 * AccessibilityFromStream :
   + InStreamType::ED_RNAplfold_Text : ED input in RNAplfold style
   + parseED_RNAplfold_text() : read ED input in RNAplfold style
   * parsePu_RNAplfold_text() : using parseRNAplfold_text()
   + parseRNAplfold_text() : generic RNAplfold Pu-styled text input
     + reduces availMaxLength if not enough data in input and outputs a user
       information
   + availMaxLength : stores the maximal window length parsed from the input,
     init = Accessibility::getMaxLength()
   + getMaxLength() : provides availMaxLength
 * AccessibilityVrna :
   * getED() :
     + accessibility constraint check
 * AccessibilityConstraint :
   + isAccessible( range )
   + isMarkedAccessible( range ) 
   + isMarkedBlocked( range )
   + isUnconstrained( range ) 
 * README.md :
   + docu for read/write of ED and Pu values for accessibility initialization

170129 Martin Mann :
 * PredictorMfe2dHeuristic* :
   * fillHybrid*() : 
     + interaction length check before energy evaluation
     
170128 Martin Mann :
 * Makefile :
   + info for testing directive
 * AccessibilityDisabled :
   * constructor :
     - defaults removed
 * AccessibilityVrna :
   * constructor :
     * argument order changed
 + AccessibilityFromStream : populates ED values from stream
   + InStreamType::Pu_RNAplfold_Text
   + parsePu_RNAplfold_Text() : parse RNAplfold unpaired probability text output
 * CommandLineParsing :
   + q/tAccFile : file/stream to read accessibilities from
   + q/tAcc : 
     + 'P' : read unpaired probability from file : RNAplfold text output
   + stream name checks now case insensitive

170127 Martin Mann :
 * SeedConstraint : 
   + maxED + getter : maximal ED value per sequence for a seed region
   * constructor ()
     + maxED
 * SeedHandler :
   * fillSeed() :
     + skip seed regions with ED > maxED
 * RnaSequence :
   * SequenceAlphabet order changed
 * OutputHandlerText :
   * add() :
     + seed ED values added
 * OutputHandlerCsv :
   + Pu, seedPu, start/end, ED, E_*, seedStart/End, seedED
   * id : replacement of colSep occurrences
 * CommandLineParsing :
   + seedMinPu + validator
   * constructor() :
     + seedMaxED argument
     * outCsvCol : + spaces to unlink ID list
   * outCsvCols_default = id1,start1,end1,id2,start2,end2,subseqDP,hybridDP,E
 * intaRNA :
   + storing of queryAcc object to avoid recomputation for different targets

170126 Martin Mann :
 * Interaction :
   + dotBracket() : converts to VRNA-like dot-bracket notation
   + dotBar() : converts to position+interaction encoding using dot and bar
   + dotSomething() : generic function to generate substrings for dotBracket()
     and dotBar()
 * test/Interaction :
   + dotBracket()
   + dotBar()
 * PredictorMfe : 
   * reportOptima() : 
     * bugfix : nextBest() called too often
 * OutputHandlerText :
   * add(range) inlined
 + OutputHandlerCsv : produces customizable CVS table output  
 * CommandLineParsing :
   * OutputMode : 
     + CSV
   + outCsvCols* + validator : argument variable + default value
   + initOutputHandler() : prints initial output
   * constructor :
     * outMode argument : support for CSV
     + outCsvCols argument
     * out argument : short option removed
   * parse() :
     + parseStyle setup to guide parsing
     - intial output moved to initOutputHandler()
     + initOutputHandler() call
     + CSV check
   * getOutputHandler() :
     + CSV
 * intaRNA : 
   + generic exception handling

170125 Martin Mann :
 * perl/*.pl : 
   * bugfix : -L setup was using -w argument

170123 Martin Mann :
 * AccessibilityVrna :
   * fillByConstraints() : bugfix of line shift for ED filling
   * fillByRNAplfold() :
   * fillByRNAup() :
     + explicit handling of Pu == 0 to trigger ED = ED_UPPER_BOUND
 * intaRNA :
   * debug exception catch message extended (debugger hint)
 * PredictorMfe2dHeuristic :
 * PredictorMfe2dHeuristicSeed :
   * traceBack() : bugfix : right boundary (j1,j2) could be inserted again
   * loop check simplified
 * CommandLineParsing :
   * fix of plfold L <= W check (only enabled if W>0)

170122 Martin Mann :
 * AccessibilityVrna :
   * fillByRNAplfold : bugfix memleak (pl return value not freed)
 * consistent use of CLEANUP() to deallocate memory allocations done via 'new'
 * intaRNA :
   * global try-catch-block now always enabled but forwards exception in debug
     mode (for debugger use)
 * PredictorMaxProb* :
 * PredictorMfe4d* :
   * clear() : bugfix memleak (4th dimension not freed)
 * RnaSequence :
   * SequenceAlphabet now 'ACGUN' to simplify checks

170120 Martin Mann :
 * Accessibility :
   + getES() : computes the ensemble energy of all local structures of a 
     subsequence under the assumption that the subsequence is part of a 
     multiloop
 * AccessibilityDisabled :
   + getES() : throws an exception since not computable
 * AccessibilityVrna :
   + EsMatrix type = triangular matrix
   + esValues : stores ES values if non-NULL
   + getES() : computed via McCaskill algorithm
   * constructor() :
     * vrnaHandler now const
     + computeES argument to trigger ES computation
     * RNAup-based ED computation also when plFoldW > seqLength
   + computeES() : computes the ES values via VRNA 3 interface
 * CommandLineParsing :
   * get*Accessibility() :
     + computeES flag = true in DEBUG mode (has to be replace by predictor
       selection based setup)
 * InteractionEnergy :
   * getES*() : now implemented based on accS*.getES()
   + getEU() abstract : energy for k unpaired bases within a multiloop
 * InteractionEnergyBasePair :
   - getES*() : implemented in superclass
   + getEU() : returns 0
 * InteractionEnergyIdxOffset :
   + getEU() forward
 * InteractionEnergyVrna :
   - getES*() : implemented in superclass
   + getEU() : returns k * VRNA.MLbase
 * ReverseAccessibility :
   + getES() forward

170119 Martin Mann :
 * CommandLineParsing : 
   + out : output stream selection
   + outStream : output stream to write to
   + argument 'outMode' : former 'out'
   * argument 'out' now sets output stream
   * getOutputStream() : returns now outStream
   * destructor()
     + deletion of outStream if file stream
   * parse() :
     + creation of file stream for output if out != STDOUT|STDERR
 * perl/*.pl :
   * adaptions to argument change

170118 Martin Mann :
 * CommandLineParsing :
   * validate_sequenceArgument() :
   * parseSequences() :
     * uses now RnaSequence::isValidSequenceIUPAC()

170117 Martin Mann :
 * RnaSequence :
   * SequenceAlphabet : now 'ACGU' only
   + SequenceAlphabetIUPAC : all IUPAC-conform nucleotide encodings
   * areComplementary() :
   * getCodeForChar() :
   * getUpperCase() :
     + checks in debug mode only
   * isAmbiguous() :
     * check reduced to 'N'
   * RnaSequence() :
     * ambiguity check uses now converted sequence
   + isValidSequenceIUPAC() : check for IUPAC conformity
 * CommandLineParsing :
   * validateSequenceAlphabet() :
     * now checks with RnaSequence::isValidSequenceIUPAC()
   + information about IUPAC sequence encoding

170111 Martin Mann :
 * CommandLineParsing :
   * IntaRNA v1 input details corrected

161230 Martin Mann :
 * CommandLineParsing :
   * parseSequencesFasta() : 
     + bugfix : was not reading final lines without newline at the end
 * OutputHandlerIntaRNA1detailed :
   * output corrected

161229 Martin Mann :
 * CommandLineParsing :
   * IntaRNA_v1_detailed init output now more like v1 output
 * PredictorMfe2dHeuristic* :
   * getNextBest() :
     * bugfix : seq1 range overlap check
 * PredictorMfe4d :
   * getNextBest() :
     * bugfix : getE() using wrong index order
 * PredictorMfe4dSeed :
   * getNextBest() :
     * bugfix : getE() using wrong index order
   * predict() :
     * bugfix : hybridE_seed resize missing
   * fillHybridE_seed() :
     + additional early check for too small interaction sites (no seed)
   * traceBack() :
     * bugfix : energy check right of seed corrected
     * bugfix : check entry existence before access
 * intaRNA :
   + exception catch + handling only in non-debug mode (eases debug)

161224 Martin Mann :
 * OutputHandler :
   + reportedInteractions : count of reported interactions
   + reported() : returns reportedInteractions
 * intaRNA :
   + setup of separator output for IntaRNA v1 output
 * OutputHandlerIntaRNA1detailed :
   + intialOutputDone : tag to store whether or not something was already done
   + prinSeparator : tag whether or not a separator is to be printed
   * constructor()
     - initial output (now part of add())
   * add() 
     + initial output for first reported interactions
     + counting of non-empty interactions
   + addSeparator() : sets printSeparator 

161222 Martin Mann :
 * perl/IntaRNA_1ui.pl :
   - seed energy restriction (only needed to mimic IntaRNA v1 behaviour)
   + output mode setup
 * CommandLineParsing :
   + OutputMode : enum for different output modes
   + PredictionMode_min/max 
   + outMode + validator : output mode argument variable
   * predMode : init via min/max enum entries
   * constructor :
     * argument values via toString() and enum entries to be consistent
     + '--out,-o' : argument to setup output mode
   * parse() :
     + IntaRNA1_detailed input output
   * getOutputHandler() :
     - IntaRNA1_detailed input output
     + switch based on outMode.val
   * getPredictor() :
     + PredictorMfe4dSeed support
 * OutputHandlerIntaRNA1detailed :
   * output corrected
 + perl/IntaRNA_up_1ui.pl : IntaRNA v1-styled interface wrapper for IntaRNA_up,
   i.e. RNAup like computation (4d space&time) including seed constraints
 + PredictorMfe4dSeed : seed-constraint incorporating full RNAup-like predictor

161221 Martin Mann :
 * RnaSequence : 
   * SequenceAlphabet += tT
   * getUpperCase() : 
     * replacement of 'T' with 'G'
 * Predictor :
   * updateOptima() : additional flag to avoid recomputation of overall energy
 * PredictorMfe2dHeuristic* :
   * fillHybridE() :
     + temporary vars to reduce full energy recomputation
 * SeedHandler :
   * fillSeed() :
     + returns now the number of valid seeds found 
 * PredictorMfe2dHeuristicSeed :
   * predict() :
     + stops computation if
       + no seed interaction possible
       + mfe without seed not < outConstraint.maxE 
 * SeedHandler : 
   * best hybrid energy now via table access (avoid recomputation)
 * Predictor*Seed :
   * bugfix : reported seed energy missed E_init term
 * OutputHandlerIntaRNA1detailed :
   * bugfix : newline missing at end of each interaction output

161219 Martin Mann :
 * src/Makefile :
   * bin file name changed to "IntaRNA"
 * .travis : bin file name changed
 + perl/IntaRNA-1ui.pl : wrapper to enable IntaRNA v1 like parameter usage
 * OutputConstraint :
   + maxE : maximal energy to be reported
 * CommandLineParsing :
   + outMaxE + validator : argument for maximal energy of reported interactions
 * PredictorMfe :
   * initOptima() : initializes with OutputConstraint.maxE
   * reportOptima() : ensures energy is below OutputConstraint.maxE
 * .gitignore : 
   + .* : all hidden files ignored
 + perl/Makefile.am
   + perl wrapper packaging and install
 * configure.ac : 
   + perl/Makefile

161216 Martin Mann :
 * README.md
   + feature comparison table started
 + configure.ac :
   * -fopenmp now always added to CXXFLAGS (also if VRNA path not specified)
 + OutputHandlerIntaRNA1detailed : output similar to IntaRNA v1.2.*
 * CommandLineParsing :
   * getOutputHandler() :
     + temporarily : init output of IntaRNA 1.* mode
     * temporarily : IntaRNA 1 output
   + getOutputStream() : stream to be passed to OutputHandler instances

161215 Martin Mann :
 * AccessibilityConstraint :
   * construction : bugfix reverse handling
 * CommandLineParsing : 
   * getSeedConstraint()
     + now with energy argument arguments to access seed lengths
     + reverses seed ranges for seq2
 * IndexRangeList :
   * List now std::list (before std::vector)
   + covers(range) + tests
   + reverse() + tests : reverses the indices and range order
 * PredictorMfe2d :
 * PredictorMfe2dSeed :
   * predict() : bugfix : throw without "new"
 * Predictor*Seed :
   * bugfix : uses now SeedHandlerIdxOffset instead of SeedHandler
   + seed handler offset setup
 * RnaSequence :
   * areComplementary() : bugfix : throw without "new"
 * SeedConstraint :
   * members not const (otherwise getter not needed)
   + getRanges*() : write access for ranges
 * SeedHandler : 
   * public functions virtual
   + offset* : index offset (needed for restricted matrices)
   * fillSeed() :
     * bugfix : sanity checks corrected
     * bugfix : index offset was missing
   * traceBack() :
     * bugfix : index offset was missing
 + SeedHandlerIdxOffset : index shift wrapper for SeedHandler

161214 Martin Mann :
 * intaRNA :
   * query/target enumeration swapped, such that seq1 == target in prediction
     and output
 * OutputConstraint : data structure defining all output constraints
   + enum ReportOverlap : 4 possibilities where overlapping is allowed
   + reportMax : max number of (suboptimal) interactions to report
   + reportOverlap : the overlap constraint for reported interactions
   + deltaE : maximal energy difference to mfe to be reported
 * Predictor :
   * reportMax/reportNonOverlapping replaced by outConstraint
     * predict()
     * initOptima()
     * reportOptima()
 * IndexRange :
   + regex : regex to match valid string encoding 
   + fromString() : parse string encoding
   + operator == ()
   + operator != ()
   * operator << () : now regex conform
 * IndexRangeList :
   + regex : regex to match valid string encoding 
   + fromString() : parse string encoding
   * operator << () : regex conform
   + operator == ()
   + operator != ()
   * insert() : 
     + no duplicated insertions
     + overlap checks
   * overlaps() : bugfix of predecessor check
   + shift() : shifts all indices by a given term and creates a new valid list
 * CommandLineParsing : 
   - outNonOverlapping : replaced by outOverlap
   - isOutNonOverlapping() : obsolete
   - isOutputReportMax() : obsolete
   + outOverlap + validator : overlap encoding
   + getOutputConstraint() : central data structure for output constraints
   - regexRangeEncoding : obsolete due to IndexRangeList::regex
   * parseRegion() : now uses IndexRangeList string constructor and shift() for
     initialization
 * general :
   + E_precisionEpsilon : variable holding the precision epsilon value for 
     energy equivalence comparisons
 * SeedConstraint :
   + range1 + getter : ranges for seed search in seq1
   + range2 + getter : ranges for seed search in seq2

161213 Martin Mann :
 * PredictorMfe2d :
   * initHybridE : 
     * code optimizations to reduce lookups
 * Predictor* :
   + initOptima()
   + updateOptima()
   + reportOptima()
     = generic optima handling
   - *Mfe() : replaced by *Optima()
   - *MaxPro() : replaced by *Optima()
 * IndexRangeList :
   + overlaps( range r ) : whether or not one of the ranges overlaps with r
   + tests
 * Predictor* :
   * reportOptima( parameters )
 * PredictorMfe :
   + reportedInteractions : index range lists to enable non-overlapping 
     interaction reporting
   + getNextBest() : abstract interface to access the next best non-overlapping
     interaction (based on reportedInteractions)
     -> implemented by all subclasses
   * reportOptima() : now screens for non-overlapping suboptimal interactions
     using getNextBest()

161212 Martin Mann :
 * InteractionEnergy* :
   + getES*() : ensemble energy of intramolecular substructures within a given
     sequence interval under the assumption that the region is part of an
     intermolecular multiloop, i.e. contains at least one intramolecular bp
     NOTE: not implemented yet, only functions set up
 * CommandLineParsing : 
   + oNumber + validator : max number of subopts to report
   + oNonOverlapping : whether or not overlapping interactions to report
   + opts_output : new output param category
   + getOutputNumber()
   + isOutputNonOverlapping()
 * intaRNA :
   + predict() calls now with getOutputNumber() and isOutputNonOverlapping()
 * Interaction :
   + compareEnergy() : checks whether a given interaction has energy higher
     than a given value
   * operator << (BasePair) now externally defined
 * Predictor* :
   * predict( + reportMax, + reportNonOverlapping ) : for multi-interaction out
 * PredictorMfe :
   * initMfe( + reportMax, + reportNonOverlapping ) : for multiple 
     overlapping interaction reporting
   - mfeInteraction : replaced by list of interactions
   + InteractionList : type of double-linked list of interactions
   + mfeInteractions : list of (overlapping) lowest energy interactions
   + traceBack() : now part of abstract interface to be used in reportMfe()
   + reportMfe() : for all interactions from list with E<0: calls traceBack()
     and adds to output handler
   * updateMfe() : 
     + lowest energy interaction list update
     + exclusion of duplicated list insertions (e.g. due to matrix recomputation)

161207 Martin Mann :
 * Interaction :
   - addInteraction() : obsolete
   + seedRange : optional seed information
   + setSeedRange() : initializes or overwrites seedRange 
 * PredictorMfe2dSeed :
 * PredictorMfe2dHeuristicSeed :
   * traceBack() : 
     + storing seed interaction information
 * OutputHandlerText : 
   * add() : 
     + seed information printed (if available)
     + interaction position printed
     + energy unit added

161120 Martin Mann :
 * autotools-init.sh :
   * sh -> bash 
   + pipefail : script fails on first failed command
 * CommandLineParsing :
   + new predictors registered
   * default predictor : heuristic
   + time/space infor for each predictor
   * bugfix: accessibility length/window check now after individual checks
 * Interaction :
   + ostream operator for Interaction::BasePair
   * ostream operator uses BasePair operator to reduce redundancy
 + SeedHandler : central handler for seed interaction and access for a given
     SeedConstraint
   + fillSeed() : computes the seed information
   + traceBackSeed() : traces a specific seed start
   + getSeedE() : the seed energy for a seed start (or INF if no seed possible)
   + getSeedLength*() : the seed length for a seed start (or 0 if not possible)
 * PredictorMfe :
   * updateMfe() : 
     - dangling end based pre-check removed (not sure if valid)
 * PredictorMfe2d :
   * interface documentation corrected
 * PredictorMfe2dSeed :
   - seed calculation methods and data structure (move to SeedHandler)
   + seedHandler (generated in constructor)
   * fillHybridE_seed() : 
     * bugfix : init of all cells was not ensured
     * sanity checks now include lower i boundary
 * PredictorMfe4d : prediction info adapted
 + PredictorMfe2dHeuristic : heuristic mfe prediction in n^2 time and space
 + PredictorMfe2dHeuristicSeed : heuristic mfe prediction in n^2 time and space
     including seed constraint  

161023 Martin Mann :
 * AccessibilityConstraint :
   - isBlocked() : renamed to isMarkedBlocked()
   * isAccessible() : now : whether or not a position is available for 
     interaction
   + isMarkedAccessible() : former isAccessible()
   + isMarkedBlocked() : former isBlocked()
 * AccessibilityDisabled :
   * getED() : checks now for constraint.isAccessible()
 * InteractionEnergy :
   + getBasePair() : generates an Interaction BasePair for given indices
   + getIndex*() : generates the computation indices for a given base pair
   + getED*() : fast access to ED of a sequence
   + isAccessible*() : checks for accessibility and non-ambiguity
   + size*() : sequence length
   * isValidInteraction() : now protected
   * isAllowedLoopRegion() : now protected
 + InteractionEnergyIdxOffset : wrapper for an InteractionEnergy that handles 
   index shifts due to a given offset. To this end, all according functions are
   overwritten, which eases formulations of Predictors.
 * Predictor :
   * energy : now InteractionEnergyIdxOffset object
 * PredictorXXX :
   - i*offset : obsolete due to energy class change
   * all offsets are now obsolete with extended InteractionEnergy interface and 
     InteractionEnergyIdxOffset energy access : this simplifies the formulation
     of the algorithms and makes debugging easier.
 - test/InteractionEnergy_test.cpp : obsolete

161020 Martin Mann :
 * configure.ac :
   + boost_regex lib registered
 * CommandLineParsing :
   + argument qRegion : query interaction regions (sorted, non-overlapping)
   + argument tRegion : target interaction regions
   + getter and validator
   + regexRangeEncoding : regex to validate range string encodings
   + parseRegion() : parses region argument
 * IndexRange :
   * constructor() :
     * default to = RnaSequence::lastPos
 * IndexRangeList :
   * push_back() : ascending check corrected
 * intaRNA :
   + predicts only for all index range combinations
 * ReverseAccessibility : 
   + getReversedIndexRange() : reverses an index range
 * PredictorMfe2dSeed :
   * bugfix : rotating seedE_rec usage not possible; reset to full 5D array
   * bugfix : index offset added 

161019 Martin Mann :
 + m4/m4_ax_cxx_compile_stdcxx.m4 : enables c11 compiler support check
 * configure.ac : 
   + ensures c11 compiler compliance via m4_ax_cxx_compile_stdcxx.m4
   * defining "_DEBUG" instead of "DEBUG" in debug mode
   - enable-c11 flag removed since obsolete (c11 now mandatory)
 * .travis : adaption to configure changes
 + easylogging++.h : v9.83 of header-only logging lib (MIT license)
 * general.h :
   + includes config.h for configure-based flags
   + IN_DEBUG_MODE : compiler flag to mark debug mode
   + includes easylogging++.h : global logging support
   + easylogging++ log file setup (disabled)
 * AccessibilityConstraint :
   + isUnconstrained() : checks whether or not a position is unconstrained
   * isBlocked() :
   * isAccessible() :
     + quick check for empty list
 * AccessibilityVrna :
   - log stream : obsolete
   * construction :
     + base pair span forwarded to fill* functions
     * fillRNAup() also used if sliding window size == 0 (since fastest)
   * fill*() :
     + timing added (debug mode only)
     + verbose log of computation mode
   * fillByConstraints() :
   * getPfScale() :
   * fillByRNAup() :
     + uses now base pair span
 * CommandLineParsing :
   - logStream : obsolete (global logging)
   + seedMaxE argument : maximal overall energy of a seed interaction
   + validate_seedMaxE()
   + seedConstraint : central seed constraint object for all predictors
   * getSeedConstraint() : returns constant reference to central object
 * InteractionEnergy :
   - getInterLoop() : replaced by getInterLeft()
   + getInterLeft() : left side loop extension (former getInterLoop())
   + getPr_danglingLeft() : probability of dangling ends on the left side
   + getPr_danglingRight() : probability of dangling ends on the right side
   * getE() :
   * getE_contributions() :
     * using getPr_dangling*() for dangling end energy computation
 * Predictor* :
   * predict() :
     + verbose log for prediction mode
     + performance logging in debug mode
 * PredictorMfe2dSeed :
   * fillSeed() :
     + over energy computation for seed now includes getE_init()
     * seed max energy boundary now via SeedConstraint.getMaxE()
   * traceBackSeed() :
     * bugfix : j2 initialization
 * SeedConstraint :
   + maxE + getter : maximal overall energy a seed is allowed to have

161017 Martin Mann :
 * CommandLineParsing :
   - seedMinBP : renamed to seedBP
   - getSeedMinBp : obsolete
   + noSeedRequired : whether or not to enforce seed constraint for interactions
   + seedBP : number of base pairs within seed
   + seedMaxUP : maximal overall number of unpaired bases in seed
   + seedMaxUPq : maximal number of unpaired bases within query's seed
   + seedMaxUPt : maximal number of unpaired bases within target's seed
   + validator_*() for arguments
   + getSeedConstraint() : provides the seed according to the arguments
   * parse() :
     + noSeedRequired setup
   * getPredictor() : predictor type depends now on noSeedRequired member setup
 * general.h :
   + E_isNotINF() : check for energy != E_INF
   + E_isINF() : check for energy == E_INF
 * InteractionEnergy :
   * getE() :
     + overall energy only computed if hybridE != E_INF
 * OutputHandlerText :
   * add() :
     + output text for empty interaction
 * PredictorMaxProb :
   * fillHybridZ() : bugfix of boundary checks
   * updateMaxProbInteraction() : bugfix : offset missing
 * PredictorMfe :
   * updateMfe() :
     * feasibility check extended with dangling end energies (can be < 0)
     * bugfix : offset missing
 * PredictorMfe2d/4d :
   * checks via E_isINF and E_isNotINF
 + PredictorMfe2dSeed : predicts interactions that fulfill the seed constraint
 + SeedConstraint : defines a seed constraint
   + bp : number of base pairs
   + maxUnpairedOverall : maximal overall number of unpaired bases
   + maxUnpaired1 : maximal number of unpaired bases in first sequence
   + maxUnpaired2 : maximal number of unpaired bases in second sequence
   + getMaxLength*() : provides maximal length of a seed given the constraints
   + ostream operator

161012 Martin Mann :
 * CommandLineParsing :
   + parseSequencesFasta() : read FASTA format from input stream
   * validate_sequenceArgument() :
     + stream/file support enabled
   * parseSequence() :
     + stream/file support enabled
 * intaRNA :
   * bugfix: queryAcc cleanup in inner loop

161012 Martin Mann :
 * AccessibilityVrna :
   * getPf* :
     + local vrna_md_t model for according window sizes
 * CommandLineParsing :
   * getOutputHandler() : + energy handler as argument
   + energyFile : file name of the VRNA energy parameter file to be used
   + validate_energyFile()
   + validateFile() : generic file existence and readability check
   + updateParsingCode() : central utility function for parsing code updates
   * parse() :
     + update of vrnaHandler after parsing
     + energFile handling
     + updateParsingCode() usage
   * validate_sequenceArgument() 
     + validateFile() usage
 * general.h
   + E_equal() : checks for (semi)equality for floating point variables
 * intaRNA : adaptions to interface changes
 * InteractionEnergy :
   + EnergyContributions : individual energy contributions for an interaction
   + getE_contributions() : provides an EnergyContributions container for a 
     given interaction
   + getE_init() : the duplex initiation energy
   + getE_interLoop() : former getInterLoopE()
   + getE_danglingLeft() : former getDanglingLeft()
   + getE_danglingRight() : former getDanglingRight()
   + getE_endLeft() : former getEndLeft()
   + getE_endRight() : former getEndRight()
   + getBestE_interLoop() : former getBestStackingEnergy()
   + getBestE_dangling() : former getBestDangleEnergy()
   + getBestE_end() : former getBestEndEnergy()
   - getInterLoopE() : renamed
   - getDangling*() : renamed
   - getEnd*() : renamed
   - getBest*Energy() : renamed
   - getBestInitEnergy() : obsolete
   * isValidInternalLoop() :
     + returns false on (i1==j1 || i2==j2)
 * InteractionEnergyBasePair :
   * adaptions to interface changes
   + getE_init() : returns -1.0
   * getE_interLoop() : returns only interior loop energies (no init)
 * InteractionEnergyVrna :
   * adaptions to interface changes
   * getE_interLoop() : returns only interior loop energies (no init)
   * getE_dangling*() : substracts end energies to get dangling end 
     contributions only (combined value from the VRNA interface)
 * OutputHandlerText :
   + energy : used energy handler for interaction computation
   * construction() : + energy argument
   * add() : 
     + provide energy contribution details
 * PredictorMaxProb :
   * fillHybridZ() :
     + explicit energy.getE_init() handling
 * PredictorMfe :
   * adaption to interface changes
 * PredictorMfe2d/4d :
   * fillHybridE() :
     + explicit energy.getE_init() handling
   * traceback() :
     * bugfix: i2/j2 index computation for offset > 0
     + explicit energy.getE_init() handling
 * VrnaHandler :
   * constructor() :
     + vrnaParamFile argument : for loading non-default energy parameters
     - max_bp_span : now in getModel()
     - window_size : now in getModel()
     * default for max_bp_span/window_size = -1
   * getModel() :
     * now creates a new model object per call to enable different parameters
       e.g. for different accessibility computations
     + max_bp_span argument
     + window_size argument
 * test/InteractionEnergyBasePair : adaption to interface changes

161007 Martin Mann :
 + PredictorMfe : generic interface for mfe predictors to reduce code redundancy
   + mfeInteraction
   + i1offset
   + i2offset
   + initMfe()
   + updateMfe()
   + min..E() : minimal energy contributions to avoid recomputation
 * PredictorMfe4d 
   * now PredictorMfe subclass
   - PredictorMfe members
   * using energy.areComplementary()
   * bugfix : single interaction energy was doubled
   * bugfix : ED-based cell computation skipping was check reversed
   * fillHybridE() :
     - energy argument was obsolete
 + PredictorMfe2d : space efficient mfe predictor using 2D matrices only
   + predict() : computes best interaction screening all possible right 
     interaction ends
   + initHybridE() : initializes the cells used for current right interaction 
     ends
     * bugfix : ED-based cell computation skipping was check reversed
   + fillHybridE() : computes all possible interactions for current right
     interaction ends and reports them to updateMfe()
   + traceBack() : recomputes interactions for the right side of the current
     interaction site and traces the details of the interaction of interest
 * PredictorMaxProb :
   * bugfix : single interaction energy was doubled
 * InteractionEnergy :
   + areComplementary() : checks two positions if complementary
 * CommandLineParsing :
   + predMode argument (mode,m)
   + validate_predMode() : checks the argument value
   * getPredictor() : returns predictor by argument setup

161006 Martin Mann :
 - PredictorMfeRNAup : renamed to PredictorMfe4d
 + PredictorMfe4d : former PredictorMfeRNAup
   * overall energies computed in updateMfeInteraction()
 * InteractionEnergy : 
   + getE() : computes the overall energy for a given interaction site and
     hybridization energy
     E = hybridE + ED1 + ED2 
         + dangleLeftE*prob(leftFree) + dangleRightE*prob(rightFree)
         + closeLeftE + closeRightE
   + getBoltzmannWeight() : the Boltzmann weight for a given energy
 * InteractionEnergy* : 
   + getEndLeft/Right() : the penalty for ending the interaction
   + getBestEndEnergy() : best energy penalty for interaction ends
 * InteractionEnergyVrna :
   + bpGC/CG : base pair type encoding for GC/CG base pair
   + isGC() : tests whether two positions resemble a GC/CG base pair
 * PredictorMaxProb :
   - getBoltzmannWeight() : now part of InteractionEnergy 
   * overall energies/weights computed in updateMaxProbInteraction()

161005 Martin Mann :
 * Predictor :
   + getMaxInteractionWidth() : upper width bound for an interacting site
 * PredictorMaxProb :
 * PredictorMfeRNAup :
   * matrix initialization of end regions reduced to the accessible window
     widths using getMaxInteractionWidth()
   * adaption of recursion to varying matrix sizes
 + test/Interaction
 * test/Makefile.am : clear binary before each test run (ensures relinking)
 * Interaction :
   * isValid() : bugfix : wrong return for single bp interaction

161005 Martin Mann :
 * Accessibility* :
   - AccessibilityConstraintAlphabet : now part of AccessibilityConstraint
   * ED_UPPER_BOUND = E_INF
   * uses now AccessibilityConstraint object instead of dot-bracket string
 + AccessibilityConstraint : encodes an accessibility constraint based
   + init via dot-bracket VRNA-like encoding
   + getVrnaDotBracket() : get according VRNA-styled dot-bracket encoding
   + isAccessible() : check for explicit "accessible" marking
   + isBlocked() : check for explicit "blocked" marking
   + isEmpty() : check if any constraint information present
 * CommandLineParsing :
   * getTargetAccessibility() : creation of AccessibilityConstraint object
   * getQueryAccessibility() : creation of AccessibilityConstraint object
 + IndexRange : represents an ascending or descending range of (sequence) indices
   + isAscending() : check if ascending range encoding
   + isDescending() : check if descending range encoding
   + ostream operator
   + less-than operator
 * Interaction :
   + construction from InteractionRange
   * sequence handling via pointers to enable assignment operator
   * basePairs : public
   - getBasePairs() : obsolete for public member
   * energy : public
   - get/setEnergy() : obsolete for public member
   + isEmpty() : check if any base pairs present
   + assignment operator (InteractionRange)
 * InteractionEnergy* :
   + getBestStackingEnergy() : access to the best stacking energy contribution
   + getBestInitEnergy() : access to the best initiation energy contribution
   + getBestDangleEnergy() : access to the best dangling end energy contribution
 + InteractionRange : represents two interacting ranges of two sequences based
   on two according IndexRange objects
   + construction from Interaction (boundaries)
   + isSane() : checks ordering of index ranges
   + ostream operator
   + less-than operator
   + assignment operator (Interaction)
 * src/Makefile.am : adaption to source changes
 * OutputHandler* :
   + add(InteractionRange)
 + OutputHandlerRangeOnly : converts reported Interactions into an 
   InteractionRange and forwards the output to a successive OutputHandler
 * OutputHandlerText :
   * adaption to Interaction interface changes
 * Predictor* :
   * predict(IndexRange,IndexRange) : takes to ranges instead of 4 coordinates
 * PredictorMaxProb :
   * maxProbInteraction : now an InteractionRange 
 * PredictorMfeRNAup :
   + getE() : computes the overall energy for a given interaction site
   * updateMfe() : takes only the overall energy instead of individual contribs
 * ReverseAccessibility :
   * adaption to AccessibilityConstraint usage
 * RnaSequence :
   + isAmbiguous() : checks a certain sequence position for ambiguity
 + test/AccessibilityConstraint
 + test/IndexRange
 + test/InteractionRange
 + test/OutputRangeOnly

161004 Martin Mann :
 * general :
   - E_MAX : replaced by E_INF
   + E_INF : infinity representation for E_type

161001 Martin Mann :
 - tests/test* : renamed to *_test
 + tests/*_test.cpp : test files for the according classes
 * tests/Makefile : 
   + wildcard based test file listing
 * travis :
   + make test added
 * configure.ac : back to automake 1.11 features
 * */Makefile.am : test target adapted to configure change
 
160930 Martin Mann :
 * InteractionEnergy* :
   + getRT() : provides the RT constant for this energy model to compute
     Boltzmann weights
 + PredictorMaxProb : computes the interaction site with maximal probability
   and reports the according boundaries and the site's ensemble energy to the
   output handler 
160930 Martin Mann :
 * PredictorMfeRNAup :
   * mfeInteraction base pairs now updated instead of recreation (reduced memory
     allocation events) 
 - PredictorRNAup.* : renamed to PredictorMfeRNAup.*
 + PredictorMfeRNAup.* : rename to be class name conform
 * Makefile.am : adaption to file rename
 * intaRNA : 
   + call CommandLineParsing.getPredictor() to get predictor
 * CommandLineParsing :
   + getPredictor() : generates the predictor according to the user request
160930 Martin Mann :
 * InteractionEnergyVrna :
   * getDangling*() : bugfix : dangling sequence code was not given
 * ReverseAccessibility :
   + getReversedIndex() : get index in reversed sequence
 * Interaction :
   + energy + get/set() : optional energy value for the interaction (init NaN)
 * Predictor :
   * output non-const
   * constructor :
     * output non-const (otherwise no filling possible)
   + predict() : triggers prediction
 - PredictorRNAup : renamed PredictorMfeRNAup
 + PredictorMfeRNAup :
   + mfeInteraction : holds the mfe interaction
   - hybridEmin : now part of mfeInteraction
   + initMfe() : initializes mfe and mfe interaction
   + updateMfe() : updated mfe and mfe interaction
   * fillHybridE() :
     * global temp variables to reduce allocation effort
     + dangling end contribution added
   * constructor() : only initializes data structures, no computation triggered
   + predict() : does the computation and output reporting
   + clear() : clears temporary data structures
   * destructor() : calls now only clear()
   + i1offset : the offset for the first sequence for index handling
   + i2offset : the offset for the second sequence for index handling
 * OutputHandlerText :
   + add() : 
     + prints interaction energy
 * RnaSequence
   + last : placeholder to represent the last position within a sequence
 * intaRNA :
   * calls predictor->predict()
 * AccessibilityVrna :
   * calc_ensemble_free_energy() 
     * input check preprocessor defines corrected

160929 Martin Mann :
 * CommandLineParsing :
   + ReturnCode : enum for different return codes for parsing
   - parsingResultNotSet : obsolete (covered by ReturnCode)
   * adaption to ReturnCode usage
 * intaRNA :
   * adaption to ReturnCode
   + PredictorRNAup temporarily hard-coded
 + unit tests added using Catch header library
   * configure.ac : 
     + generate tests/Makefile
     + AM_EXTRA_RECURSIVE_TARGETS([tests]) : recursive "tests" make target
     * AM_INIT_AUTOMAKE([1.13]) : need v1.13 for AM_EXTRA_RECURSIVE_TARGETS
   * Makefile.am :
     + go to subdir 'tests' for 'tests' make directive only
   + tests/Makefile.am : compile and run tests 
   + catch.hpp : Catch library version 1.5.7
   + tests/runTests.cpp : central test program
   + tests/testInteractionEnergy.cpp 
   + tests/testInteractionEnergyBasePair.cpp 
   + tests/testRnaSequence.cpp
 * RnaSequence :
   + areComplementary() : tests for base pair complementarity
   * constructor : input checks in debug mode
 * Accessibility + subclasses :
   * constructor : accConstraint now pointer to enable NULL argument
 * CommandLineParsing :
   * argument docu extended
   * adaption to Accessibility interface change
   + energy argument added
   + validate_energy()
 * InteractionEnergy :
   * energy getters now const
   * isAllowedLoopRegion() 
     * now public
     * bugfix of maxInternalLoppSize check 
   + getMaxInternalLoopSize*() to access individual max internal loop size
   + isValidInternalLoop() : checks an internal loop request
 * InteractionEnergyVrna :
   * energy getters : bugfix normalization dcal to kcal
   * interface adaption
   * getInterLoopE() : 
   * getDanglingLeft() : 
   * getDanglingRight() : 
     * bugfix of sequence access
 * InteractionEnergyBasePair :
   * interface adaption
 * PredictorRNAup :
   * E2dMatrix : now quadratic matrix for interaction widths w1,w2
   * E4dMatrix : now quadratic matrix for interaction starts i1,i2
   + hybridEmin : global minimal energy value computed
   * (de/con)structor : matrix deletion/initialization adapted
   * fillHybridE() :
     + two round computation: first hybridizations only, than ED values added

160928 Martin Mann :
 * AccessibilityVrna :
   + fillByConstraints() : intarna-1 way of computing ED values
   + fillByRNAup() : RNAup-like ED value computation
   + fillByRNAplfold() : RNAplfold-like ED value computation
   * remaining interface adapted to local computation and data structures
   * using boost::banded_matrix for ED storage
   * constructor() :
     + ED setup depends on whether or not structure constraint present
   * getPfScale() : bugfix : now use global accessibility constraint for 
     mfe/ensemble energy computation
   * Accessibility member access now via getter
   * adaption to possibly empty accConstraint
 * Accessibility :
   * constructor() : accConstraint now empty if not provided to reduce memory
   * checkIndices() : checks only in DEBUG mode enabled
 * CommandLineParsing :
   * cleanup 
 * intaRNA :
   * adaption to parse return value change
 * InteractionEnergyVrna :
   * destructor : bugfix : has to use 'free' for VRNA cleanup
 * VrnaHandler :
   + getRT() : computes the RT constant for the current temperature
 * configure.ac :
   * VRNA check extended
 + PredictorRNAup : RNAup-like interaction prediction
 * Makefile.am : adaption to source file changes
 * configure.ac :
   + additional check for "-std=c++11" support of the compiler
 * InteractioNEnergy :
   + getAccessibility* : access to full accessibility objects
   - getLength* : obsolete

160927 Martin Mann :
 + AccessibilityVrna : vrna-based accessibility computation 
   (former AccessibilityVienna)
 + InteractionEnergy : former Energy
 + InteractionEnergyBasePair : former EnergyBasePair
 + InteractionEnergyVrna : former EnergyVienna
 - AccessibilityVienna : renamed 
 - Energy : renamed
 - EnergyBasePair : renamed
 - EnergyVienna : renamed
 * Accessibility :
   + ostream operator : prints ED values
 * AccessibilityVrna :
   - modelDetails : obsolete
 * RnaSequence :
   + ostream operator : prints "ID(Sequence)"
 + VrnaHandler : central handler for VRNA folding parameter setup
   * model generation now done in constructor
   * model access non-const due to VRNA interface ("features" on the fly 
     initialization of model if not initialized yet, thus non-const)
 * CommandLineParsing : 
   + vrnaHandler
   + getTemperature() : getter for user defined temperature
   * adaption to renames
 

160923 Martin Mann :
 * .travis :
   * sudo: true (was false)

160922 Martin Mann :
############### MOVE FROM CVS TO GIT INTARNA REPO ############################
 * README : 
   + compilation information
   + refers to README.md
 + README.md (by Bjoern Gruening)
 * COPYING : refers to LICENSE
 + LICENSE : now MIT license for intarna 2.*
 * configure.ac : replaced by new version for intarna 2.*
 .travis.yml : updated to Vienna 2.2.10 usage 
 + m4/m4_ax_boost_base.m4 : boost m4 checks for configure.ac usage
 + autotools-init.sh : calls autotools to generate system specific files
   ==> NOTE : CALL AFTER CHECKOUT <==
 - depcomp : temporary autotools file generated by 'bash autotools-init.sh'
 - install-sh : temporary autotools file generated by 'bash autotools-init.sh'
 - configure : temporary autotools file generated by 'bash autotools-init.sh'
 - missing : temporary autotools file generated by 'bash autotools-init.sh'
 - Makefile.in : temporary autotools file generated by 'bash autotools-init.sh'
 - aclocal.m4 : temporary autotools file generated by 'bash autotools-init.sh'
 - src/*.h : old source removed
 - src/*.c* : old source removed
 - src/data/ : old energy data removed
 + src/*.h : new source data for version 2.*
 + src/*.cpp : new source data for version 2.*
 * src/Makefile.am : updated for new source files

160921 Martin Mann :
 * RnaSequence.h : documentation corrected
 * CommandLineParsing :
   + validateSequenceAlphabet() : checks 
     + sequence alphabet check
     + ambiguity check for warning
   * parseSequences() :
     + validateSequenceAlphabet() call

150616 Martin Mann
 * CommandLineParsing :
   * parse() 
     + check for minimal sequence length (>= seedMinBP)

140719 Martin Mann
 * EnergyBasePair : 
   * getInterLoopE() : BUGFIX : was returning loop energy instead of negated 
     base pair number only

140716 Martin Mann
 + OutputHandlerText : prints reported interaction in a simple text format to
   stream similar to IntaRNA 1.* output
   
140714 Martin Mann
 * general : 
   + CHECKNOTNULL macro to throw an exception with message if a given pointer is
     not initialized
 + Predictor : general interaction prediction interface
 + Interaction : interaction information container to be filled by predictors 
   and to be handled by output handlers
 + OutputHandler : general interaction reporting handler interface

140703 Martin Mann
 * Accessibility* :
   + global accessibility constraint support (some sequence positions have to be
     unstructured both in structure ensemble and interaction and are thus 
     omitted from interaction prediction via ED_UPPER_BOUND accessibility 
     values)
 * CommandLineParsing :
   + *AccConst : parameter + checks for accessibility constraints
   + validateSequenceNumber() : sequence number check (for depending parameters)

140701 Martin Mann
 * intaRNA :
   + general computation setup started
 * general :
   + CLEANUP : macro to delete a pointer and to set it to NULL
 * CommandLineParsing :
   + CharParameter : ensures a char parameter is from a given alphabet
   + validate_charArgument() : generic check function
   + validate_numberArgument() : generic check function
   + several parameters including check functions, defaults, etc.
   + get*Accessibility() : generates an Accessibility object according to the 
     user parameters
   + getEnergyHandler() : generates an Energy object according to the parameters
 * ReverseAccessibility :
   + access to original sequence and accessibility object
 * RnaSequence :
   + ambiguity check

140630 Martin Mann
 + ReverseAccessibility : reversed index access to a sequence and its 
   accessibility (ED) values as needed for the energy computation
 * AccessibilityVienna :
   * Vienna RNA package calls corrected
   + partFoldParams : central partition function folding parameter object 
 + Energy : abstract interface to compute energy contributions
   + getDangling*() : dangling end penalties
 + EnergyBasePair : energy computation in terms of number of base pairs
 + EnergyVienna : energy computation via Vienna RNA package

140627 Martin Mann
 * Accessibility* :
   + maxLength : maximal length of accessible regions to consider, defaults to 
     the sequence's length
 + E_type : global type for energy variables/values
 + T_type : global type for temperature variables/values
 + E_MAX : constant : maximal value for energies

140625 Martin Mann
 * configure :
   + Vienna RNA package v2.* path support and checks
 + Accessibility : abstract interface to get accessibility energies for regions
 + AccessibilityNone : Accessibility implementation that always returns zero
 + AccessibilityVienna : Accessibility implementation that computes _all_ ED 
   values based on ensemble energies computed via the Vienna RNA package
 * RnaSequence :
   + as* : access functions for different sequence encodings

140624 Martin Mann
 + general.h :
   + macro to stop at unimplemented source code locations via std::runtime_error 
 + RnaSequence : class that represents an RNA sequence with according 
   functionality
   + integer encoding
   + string validation
 * CommandLineParser :
   + query and target : 
     + support for direct sequence input
     + preparation for stream/file parsing

140618 Martin Mann
 + cygwin boost installation using
     ./b2 --prefix=/cygdrive/c/Research/Software/boost/1.55.0  --layout=tagged
        --without-python --without-mpi --without-context threading=single 
        link=static variant=release install
 + src/CommandLineParsing : class that handles command line argument parsing
   and checking via boost::program_options
   NOTE: under cygwin a boost build from source is needed to get the libraries
         correctly compiled and linkable (boost lib from cygwin repos is not
         working)

140617 Martin Mann
 + autoconf setup

140617 ********  start reimplementation of IntaRNA  ******************


#############################################################################

2011-06-10 Andreas Richter      version 1.2.5
   * fixed bug in string clearing when option -v is used and no
     significant interactions are found
   * fixed bug in parameter checking

2011-03-24 Andreas Richter      version 1.2.4
   * changes in error messages

2010-08-13 Andreas Richter	version 1.2.3
   * fixed bug in computation of C matrix for parameter -n

2009-10-08 Andreas Richter
   * added support for ambiguous nucleotide(s) "N"

2009-09-11 Andreas Richter      version 1.2.2
   * fixed bug in automated setting the maximal length of unpaired region in
     target

2009-09-07 Andreas Richter
   * fixed bug in range check for parameter -w (disallow values <3)
   * fixed bugs in calculation of ED values of target RNA and binding RNA with
     RNAplfold and RNAup, respectively, concerning pf_scale for RNAplfold and in
     combination with values 0 for parameter -a/-b (weights for ED values) 

2009-06-11 Andreas Richter
   * threshold for seed accessibility added as EXPERIMENTAL FEATURE

2009-06-11 Andreas Richter      version 1.2.1
   * fixed bug of wrong seed region positions when options -f, -m and -t are used together
   * fixed bug in range check of parameter -f

2009-05-29 Andreas Richter
   * weight for ED values of target RNA and binding RNA in energy added

2009-02-23 Andreas Richter      version 1.2
   * new default parameters (-p 6, -U, -P)
   * ED calculation of target RNA now by new RNAplfold of Vienna Package 1.8.x

2008-12-10 Andreas Richter      version 1.1.1
   * ED calculation of small RNA by RNAup (-U option) added

2008-10-31 Anke Busch           version 1.1
   * bug in ED calculation of small RNA fixed

2008-10-24 Andreas Richter
   * bug in calculation of suboptimal solutions fixed

2008-10-02 Andreas Richter
   * new parameter -f that allows to defined search region for seed
   * more efficient calculation of matrices
   * minor bug in dangling end contributions fixed

2008-09-03 Andreas Richter
   * temperature-dependent calculation of interaction implemented

2008-08-04 Andreas Richter
   * intermolecular initiation free energy added

2008-07-16 Andreas Richter
   * complete approach without heuristic (-n option) added

2008-07-11 Andreas Richter
   * seed calculation corrected (C matrix indices)

2008-07-08  Andreas Richter
   * dangling end energy contributions corrected

2008-07-03  Andreas Richter
   * integrity check for -u parameter added
   
2008-06-11  Anke Busch          version 1.0.0