- Python 3.7 port, removed support for Python 2.x
- Support for Vcf4Records and merging VCF files
- Better handling of h5py
- Support 4.0 basecaller and chemistry
- Misc fixes
- Support for 3.2 basecaller and chemistry
- GFF bug fixes
- Misc fixes and enhancements
- Update to pysam 0.9.0
- Recognize new and prospective future partnumbers for the 3.1 timeframe
- Misc fixes for speeding up dataset and cmp.h5 access
- Rename pulseFeature accessors to “baseFeature…”, since they reflect features accompanying basecalls as opposed to pulse calls; add “pulseFeature…” accessors that actually get the pulse features (from an internal-mode BAM)
- Improve FastaWriter performance with IndexedFastaRecords
- Improve DataSet chunking, filtration, consolidation
- Dataset improvements
- Dataset CLI moved to pbcoretools
- Support for context flag filtering
- Misc fixes
- Improved IPython tab completion
- Can now use .open on Dataset XML files
- Update to pysam 0.8.4
- Miscellaneous bugfixes
- Recognize sequencing kit C4v2
- Support for multipart bas files with the new Sequel holenumber format
- Add readsByHoleNumber to aln file classes
- Accessors for SNR from BAM reads; needed for Arrow access
- add GFF gather functionality
- support for 3.0.1 BAM format spec
- Updated for newer dataset schema
- BAM access speed improvements
- BGZF fixed for large bam.pbi files
- readsInRange iteration order parity between different dataset sources
- Bug fixes and speed improvements to the
- Dataset constructor APIs enhanced
- Same as 1.2.0; first version on PyPI
- Support for the new BGZF-based bam.pbi file format
- Support for PacBio BAM spec version 3.0b7; previous versions are rejected
- Added dataset functionality to pbcore.io
- Added dataset.py tools for CLI manipulation of DataSet XML files
- BAM fixes
- Better FOFN handling
- Recognize additional part numbers for P6 chemistry
- Better –debug in ToolRunner
- BAM fixes
- Adopt aStart, aEnd conventions from BAM spec, replacing rStart, rEnd
- AlnIndex sharing among CmpH5Reader objects
- BAM support moved to 3.0 spec; support for earlier PacBio BAMs dropped
- Deprecation warning decorators added
- BAM-incompatible cmp.h5 accessors deprecated
- Moved to support 3.0 FASTA conventions in Fasta readers
- Rename FastaTable to IndexedFastaReader
- “.open” script added for convenience
- openers added (factory methods invoking the appropriate Reader class; useful for applications that want to transparently use either BAM or cmp.h5)
- BAM support: Addition of BamReader, IndexedBamReader, and BamAlignment
- Minor CmpH5Reader API changes for greater compatibility with BamReader (deprecation of movieInfo in favor of readGroupInfo)
- Removed unused components from CmpH5Reader API
- Add example BAM file
- Length accessors for FAST[AQ] record types
- FASTA header parsing into “id” and “metadata” now available in the FastaRecord types
- pbcore.chemistry: a new subpackage for decoding barcode information to the human-readable chemistry name
- BasH5Reader: more robust handling of broken region tables
- CmpH5Reader: loading an empty cmp.h5 will raise an EmptyCmpH5Error. This is because the semantics of an empty cmp.h5 were never defined, and for example it is not defined whether or not a cmp.h5 lacking a movie table is compliant.