### Enhancements
- **Rope-based parsing**: Added direct rope parsing for FASTA, EMBL, and FASTQ formats via `FastaChunkParserRope`, `EmblChunkParserRope`, and `FastqChunkParserRope`. Sequence extraction now supports U→T conversion and improved line ending detection.
- **Rope scanner refactoring**: Unified rope scanning logic under a new `ropeScanner`, improving maintainability and consistency.
- **Sequence handling**: Added `TakeQualities()` method to BioSequence for more efficient quality data handling.
### Bug Fixes
- **Compression behavior**: Fixed `CompressStream` to correctly use the `compressed` variable instead of a hardcoded boolean.
- **String splitting**: Replaced ambiguous `SplitInTwo` calls with precise `LeftSplitInTwo` or `RightSplitInTwo`, and added dedicated right-split utility.
### Tooling & Workflow Improvements
- **Makefile enhancements**: Added colored terminal output, a `help` target for documenting all targets, and improved release workflow automation.
- **Release process**: Refactored `jjpush` into modular targets (`jjpush-notes`, `jjpush-push`, `jjpush-tag`), replaced `orla` with `aichat` for AI-assisted release notes, and introduced robust JSON parsing using Python. Release notes are now generated and stored in temp files for tag creation.
- **Versioning**: `bump-version` now supports the VERSION environment variable for manual version setting.
- **Submission**: Switched from raw `jj git push` to `stakk` for PR submission.
### Internal Notes
- Installation instructions are now included in release tags.
- Fixed-size carry buffer replaced with dynamic slice for arbitrarily long line support without extra allocations.
Replace SplitInTwo calls with LeftSplitInTwo or RightSplitInTwo depending on the intended split direction. In fastseq_json_header.go, extract rank from suffix without splitting; in biosequenceslice.go and taxid.go, use LeftSplitInTwo to split from the left; add RightSplitInTwo utility function for splitting from the right.
Replace the fixed [256]byte carry buffer with a dynamic []byte slice to support arbitrarily long lines without heap allocation during accumulation. Update all carry buffer handling logic to use len(s.carry) and append instead of fixed-size copy operations.
Introduce EmblChunkParserRope function to parse EMBL chunks directly from a rope without using Pack(). Add extractEmblSeq helper to scan sequence sections and handle U to T conversion. Update parser logic to use rope-based parsing when available, and fix feature table handling for WGS entries.
This commit refactors the rope scanner implementation by renaming gbRopeScanner to ropeScanner and extracting the common functionality into a new file. It also introduces a new FastqChunkParserRope function that parses FASTQ chunks directly from a rope without Pack(), enabling more efficient memory usage. The existing parsers are updated to use the new rope-based parser when available. The BioSequence type is enhanced with a TakeQualities method for more efficient quality data handling.
Introduce FastaChunkParserRope for direct rope-based FASTA parsing, enhance sequence extraction with whitespace skipping and U->T conversion, and update parser logic to support both rope and raw data sources.
- Added extractFastaSeq function to scan sequence bytes directly from rope
- Implemented FastaChunkParserRope for rope-based parsing
- Modified _ParseFastaFile to use rope when available
- Updated sequence handling to support U->T conversion
- Fixed line ending detection for FASTA parsing
Replace NewBioSequence with NewBioSequenceOwning in genbank_read.go to take ownership of sequence slices without copying, improving performance. Update biosequence.go to add the new TakeSequence method and NewBioSequenceOwning constructor.
Ajout d'un parseur GenBank basé sur rope pour réduire l'usage de mémoire (RSS) et les allocations heap.
- Ajout de `gbRopeScanner` pour lire les lignes sans allocation heap
- Implémentation de `GenbankChunkParserRope` qui utilise rope au lieu de `Pack()`
- Modification de `_ParseGenbankFile` et `ReadGenbank` pour utiliser le nouveau parseur
- Réduction du RSS attendue de 57 GB à ~128 MB × workers
- Conservation de l'ancien parseur pour compatibilité et tests
Réduction significative des allocations (~50M) et temps sys, avec un temps user comparable ou meilleur.
Implémente une optimisation du parsing des grandes séquences en évitant l'allocation de mémoire inutile lors de la fusion des chunks. Ajoute un support pour le parsing direct de la structure rope, ce qui permet de réduire les allocations et d'améliorer les performances lors du traitement de fichiers GenBank/EMBL et FASTA/FASTQ de plusieurs Gbp. Les parseurs sont mis à jour pour utiliser la rope non-packée et le nouveau mécanisme d'écriture in-place pour les séquences GenBank.
This commit fixes an issue in the GenBank parser where empty parts were being included in the parsed data. It also introduces a new script `release_notes.sh` to automate the generation of GitHub-compatible release notes for OBITools4 versions, including support for LLM summarization and various output modes.
Add support for .gbff and .gbff.gz file extensions in sequence reader.
Update the logic to return an error instead of using NilIBioSequence when no sequence files are found, improving the error handling and user feedback.
Refactor k-mer matching to use a pipeline architecture with improved concurrency and memory management:
- Replace sort.Slice with slices.SortFunc and cmp.Compare for better performance
- Introduce PreparedQueries struct to encapsulate query buckets with metadata
- Implement MergeQueries function to merge query buckets from multiple batches
- Rewrite MatchBatch to use pre-allocated results and mutexes instead of map-based accumulation
- Add seek optimization in matchPartition to reduce linear scanning
- Refactor match command to use a multi-stage pipeline with proper batching and merging
- Add index directory option for match command
- Improve parallel processing of sequence batches
This refactoring improves performance by reducing memory allocations, optimizing k-mer lookup, and implementing a more efficient pipeline for large-scale k-mer matching operations.
This commit introduces entropy-based filtering for k-mers to remove low-complexity sequences. It adds:
- New KmerEntropy and KmerEntropyFilter functions in pkg/obikmer/entropy.go for computing and filtering k-mer entropy
- Integration of entropy filtering in the k-mer set builder (pkg/obikmer/kmer_set_builder.go)
- A new 'filter' command in obik tool (pkg/obitools/obik/filter.go) to apply entropy filtering on existing indices
- CLI options for configuring entropy filtering during index building and filtering
The entropy filter helps improve the quality of k-mer sets by removing repetitive sequences that may interfere with downstream analyses.
This commit introduces sparse index support for KDI files to enable fast random access during k-mer matching. It adds a new .kdx index file format and updates the KDI reader and writer to handle index creation and seeking. The changes include:
- New KdxIndex struct and related functions for loading, searching, and writing .kdx files
- Modified KdiReader to support seeking with the new index
- Updated KdiWriter to create .kdx index files during writing
- Enhanced KmerSetGroup.Contains to use the new index for faster lookups
- Added a new 'match' command to annotate sequences with k-mer match positions
The index is created automatically during KDI file creation and allows for O(log N / stride) binary search followed by at most stride linear scan steps, significantly improving performance for large datasets.
Refactor lowmask options to use shared kmer options and CLI getters
This commit refactors the lowmask subcommand to use shared kmer options and CLI getters instead of local variables. It also moves the kmer size and minimizer size options to a shared location and adds new CLI getters for the lowmask options.
- Move kmer size and minimizer size options to shared location
- Add CLI getters for lowmask options
- Refactor lowmask to use CLI getters
- Remove unused strings import
- Add MaskingMode type and related functions
Add a new boolean option 'keep-shorter' to preserve fragments shorter than kmer-size during split/extract mode.
This change introduces a new flag _lowmaskKeepShorter that controls whether fragments
shorter than the kmer size should be kept during split/extract operations.
The implementation:
1. Adds the new boolean variable _lowmaskKeepShorter
2. Registers the command-line option "keep-shorter"
3. Updates the lowMaskWorker function signature to accept the keepShorter parameter
4. Modifies the fragment selection logic to check the keepShorter flag
5. Updates the worker creation to pass the global flag value
This allows users to control the behavior when dealing with short sequences in
split/extract modes, providing more flexibility in low-complexity masking.
This commit introduces max frequency filtering to limit k-mer occurrences and adds functionality to save the N most frequent k-mers per set to CSV files. It also includes the ability to output k-mer frequency spectra as CSV and updates the CLI options accordingly.
This commit refactors the k-mer index management tools to use a unified subcommand structure with obik, adds support for per-set metadata and ID management, enhances the k-mer set group builder to support appending to existing groups, and improves command-line option handling with a new global options registration system.
Key changes:
- Introduce obik command with subcommands (index, ls, summary, cp, mv, rm, super, lowmask)
- Add support for per-set metadata and ID management in kmer set groups
- Implement ability to append to existing kmer index groups
- Refactor option parsing to use a global options registration system
- Add new commands for listing, copying, moving, and removing sets
- Enhance low-complexity masking with new options and output formats
- Improve kmer index summary with Jaccard distance matrix support
- Remove deprecated obikindex and obisuperkmer commands
- Update build process to use the new subcommand structure
Refactor k-mer index building to use the new disk-based KmerSetGroupBuilder instead of the old KmerSet and FrequencyFilter approaches. This change introduces a more efficient and scalable approach to building k-mer indices by using partitioned disk storage with streaming operations.
- Replace BuildKmerIndex and BuildFrequencyFilterIndex with KmerSetGroupBuilder
- Add support for frequency filtering via WithMinFrequency option
- Remove deprecated k-mer set persistence methods
- Update CLI to use new builder approach
- Add new disk-based k-mer operations (union, intersect, difference, quorum)
- Introduce KDI (K-mer Delta Index) file format for efficient storage
- Add K-way merge operations for combining sorted k-mer streams
- Update documentation and examples to reflect new API
This refactoring provides better memory usage, faster operations on large datasets, and more flexible k-mer set operations.
Refactor kmer index package to use disk-based partitioning with minimizer
- Replace roaring64 bitmaps with disk-based kmer index
- Implement partitioned kmer sets with delta-varint encoding
- Add support for frequency filtering during construction
- Introduce new builder pattern for index construction
- Add streaming operations for set operations (union, intersect, etc.)
- Add support for super-kmer encoding during construction
- Update command line tool to use new index format
- Remove dependency on roaring bitmap library
This change introduces a new architecture for kmer indexing that is more memory efficient and scalable for large datasets.
This commit optimizes the low-complexity masking algorithm by:
1. Precomputing logarithm values and normalization tables to avoid repeated calculations
2. Replacing the MinMultiset-based sliding minimum with a more efficient deque-based implementation
3. Improving entropy calculation by using precomputed n*log(n) values
4. Simplifying the circular normalization process with precomputed tables
5. Removing unused imports and log statements
The changes significantly improve performance while maintaining the same masking behavior.
This commit introduces a new configuration module `obidefault` to manage progress bar settings, allowing users to disable progress bars via a `--no-progressbar` option. It updates various packages to conditionally display progress bars based on this new configuration, improving user experience by providing control over progress bar output. The changes also include improvements to progress bar handling in several packages, ensuring they are only displayed when appropriate (e.g., when stderr is a terminal and stdout is not piped).
This commit fixes an issue in the chunk processing logic where the wrong classifier instance was being reset and used for code generation. A local clone of the classifier is now created and used to ensure correct behavior during dereplication.
Modify CLIProgressBar function to check if stdout is a named pipe and disable the progress bar accordingly. This prevents the progress bar from being displayed when the output is redirected or piped to another command.
This commit addresses a bug in the super k-mer implementation where the minimizer bijection property was not properly enforced. The fix ensures that:
1. All k-mers within a super k-mer share the same minimizer
2. Identical super k-mer sequences have the same minimizer
The changes include:
- Fixing the super k-mer iteration logic to properly validate the minimizer bijection property
- Adding a comprehensive test suite (TestSuperKmerMinimizerBijection) that validates the intrinsic property of super k-mers
- Updating the .gitignore file to properly track relevant files
This resolves issues where the same sequence could be associated with different minimizers, violating the super k-mer definition.
This commit adds the implementation of the obisuperkmer command, including:
- The main command in cmd/obitools/obisuperkmer/
- The package implementation in pkg/obitools/obisuperkmer/
- Automated tests in obitests/obitools/obisuperkmer/
- Documentation for the implementation and tests
The obisuperkmer command extracts super k-mers from DNA sequences, following the standard OBITools architecture. It includes proper CLI option handling, validation of parameters, and integration with the OBITools pipeline system.
Tests cover basic functionality, parameter validation, output format, metadata preservation, and file I/O operations.
This commit refactors the SuperKmer extraction functionality to use Go's new iterator pattern. The ExtractSuperKmers function is now implemented as a wrapper around a new IterSuperKmers iterator function, which yields results one at a time instead of building a complete slice. This change provides better memory efficiency and more flexible consumption of super k-mers. The functionality remains the same, but the interface is now more idiomatic and efficient for large datasets.
Add Jaccard distance and similarity computations for KmerSet and KmerSetGroup
This commit introduces Jaccard distance and similarity methods for KmerSet and KmerSetGroup.
For KmerSet:
- Added JaccardDistance method to compute the Jaccard distance between two KmerSets
- Added JaccardSimilarity method to compute the Jaccard similarity between two KmerSets
For KmerSetGroup:
- Added JaccardDistanceMatrix method to compute a pairwise Jaccard distance matrix
- Added JaccardSimilarityMatrix method to compute a pairwise Jaccard similarity matrix
Also includes:
- New DistMatrix implementation in pkg/obidist for storing and computing distance/similarity matrices
- Updated version handling with bump-version target in Makefile
- Added tests for all new methods
This commit refactors the k-mer normalization functions, renaming them from 'NormalizeKmer' to 'CanonicalKmer' to better reflect their purpose of returning canonical k-mers. It also introduces new quorum operations (AtLeast, AtMost, Exactly) for k-mer set groups, along with comprehensive tests and benchmarks. The version commit hash has also been updated.
This commit translates all French comments in the kmer filtering and set management code to English, improving code readability and maintainability for international collaborators.
Ajout de la fonctionnalité de sauvegarde et de chargement pour FrequencyFilter en utilisant le KmerSetGroup sous-jacent.
- Nouvelle méthode Save() pour enregistrer le filtre dans un répertoire avec formatage des métadonnées
- Nouvelle méthode LoadFrequencyFilter() pour charger un filtre depuis un répertoire
- Initialisation des métadonnées lors de la création du filtre
- Optimisation des méthodes Union() et Intersect() du KmerSetGroup
- Mise à jour du commit hash
This commit refactors all k-mer encoding and normalization functions to consistently use 'canonical' instead of 'normalized' terminology. This includes renaming functions like EncodeNormalizedKmer to EncodeCanonicalKmer, IterNormalizedKmers to IterCanonicalKmers, and NormalizeKmer to CanonicalKmer. The change aligns the API with biological conventions where 'canonical' refers to the lexicographically smallest representation of a k-mer and its reverse complement. All related documentation and examples have been updated accordingly. The commit also updates the version file with a new commit hash.
