v0.38

overall

• More bumping of log4j against log4shell exploit.

v0.37

overall

• Adding support for parsing HGVS repeats.
• Adding support for "chr2acc" chromosome to accession mapping.
• Bumping log4j version against log4shell exploit.

v0.36

Highlight: Pre-Built Databases

We now provide pre-built databases via Zenodo.
This addresses various issues that were caused by changing upstream data URLs.
See README.md for details.

overall

• Switching to Github Workflows for continuous integration.
• Bumping a couple of dependencies.
• Switching to JUNIT 5.
• Pre-built databases are now available via Zenodo (see README.md for instructions).
  We provide shell scripts instead of using Java programs for downloading the files.

jannovar-core

• Apply fix for (#498, PR #499 by @roland-ewald of @limbus-medtec).
  This fixes a problem with right-shifting deletions on amino acid sequences.
  See the tickes and merge request for details.
• Fixing problem with GRCh38 RefSeq annotation / projection (#524).

jannovar-cli

• Adding command vardb-import for importing annotations from VCF files into H2 database files.
• Adding command vardb-list for listing annotation meta data from vardb-import.
• Adding command vardb-annotate for annotating VCF files from Jannovar H2 database files.

jannovar-vardbs

• Deprecating previous content of the package.
  The deprecated classes are due to be removed in v0.36.
• Adding modules for importing VCF files into H2 database files, listing the meta data conents, and annotating VCF files.

jannovar-hgvs

• Fixing parsing of unchanged (=) for nucleic acide sequences (#493).
• Fixing issue with HGVS delins with del bases but no ins bases (#491).

v0.35

jannovar-core

• Update Genotype internals to pre-calculate values.

jannovar-cli

• Fixing default_sources.ini file.

v0.34

jannovar-cli

• disabeling ensemble for mouse (does not work with an hgnc file)
• updating broken links in download source file
• Adding faMT genomes for all refseq annotations
• bumping dependency on lombok

jannovar-core

• Remove hard-coded chrMT renaming. Filenames for download that have a ? in their URL will be splitted and only the first part before the ? is used a file name in the download path.
• Making faMT annotation for refSeq optional
• Update RefSeq parser that does not run into null-pointer exceptions on mouse and rat genome (e.g. when no exon defines the gene name)
• Fixing issue with block substitutions (#475).
• Fixing RefSeq build to properly assign transcript model (use best match with alignment) in the case of duplicates.
• Fixing issue in projection in the case of leading gaps (has no effect on CDS position prediction).
• Adding TranscriptModel.getTrimmedSequence() that removes leading and trailing (unaligned in RefSeq) sequence.

jannovar-htsjdk

• Bumping HTSJDK dependency to v2.20.3 because 2.20.0 has a bug in the VariantContextBuilder

v0.34

jannovar-cli

• disabeling ensemble for mouse (does not work with an hgnc file)
• updating broken links in download source file
• Adding faMT genomes for all refseq annotations
• bumping dependency on lombok

jannovar-core

• Remove hard-coded chrMT renaming. Filenames for download that have a ? in their URL will be splitted and only the first part before the ? is used a file name in the download path.
• Making faMT annotation for refSeq optional
• Update RefSeq parser that does not run into null-pointer exceptions on mouse and rat genome (e.g. when no exon defines the gene name)
• Fixing issue with block substitutions (#475).
• Fixing RefSeq build to properly assign transcript model (use best match with alignment) in the case of duplicates.
• Fixing issue in projection in the case of leading gaps (has no effect on CDS position prediction).
• Adding TranscriptModel.getTrimmedSequence() that removes leading and trailing (unaligned in RefSeq) sequence.

jannovar-htsjdk

• Bumping HTSJDK dependency to v2.20.3 because 2.20.0 has a bug in the VariantContextBuilder

v0.33

overall

• Fixing ANN annotation field to have ./unbound cardinality.
• Bumping several dependencies, including HTSJDK.

manual

• Adding conda support in installation documentation
• Fixing broken link in quickstart

jannovar-core

• Fixing annotation of SVs that look like sequence variants (#456).
  Interpretation is to use the sequence variant annotation code now.
  This fixes a bug with annotating latest ClinVar for GRCh38.
• Prevent Annotation.{getPutativeImpact,getPutativeImpact}() from returning null (#458).
• Correctly parsing RefSeq mitochondrial transcripts.
  Bumping the required versin to 0.33-SNAPSHOT to higlight this (#463).

v0.32

overall

• Changing log4j version to 2.11.2
• Changing slf4j version to 1.7.25

jannovar-cli

• Adding a simple REST server for annotating single variants.
  • Launch with jannovar-cli rest-server -d data/hg19_refseq.ser -d data/hg19_ensembl.ser
  • Then, query with /annotate-var/refseq/hg19/chr7/140453136/A/T or /annotate-var/ensembl/hg19/chr7/140453136/A/T

v0.31

This release fixes an important issue with RefSeq transcripts that align with gaps to the genome (issues #447 #450). Overall 500 genes are affected including MYH7! You probably want to upgrade to this release.

jannovar-core

• Introducing classes for representation of gapped sequences, alignments, and position projection.
• Fixing bug in ENSEMBL transcript database generation (tx version was appended twice)
• Adding flags for "has substitutions" and "has indels" to TranscriptModel that get filled for RefSeq transcripts.
• Correctly parsing of RefSeq transcripts with indels.

jannovar-cli

• Using HTTP protocol instead of FTP everywhere as it's possible (#451).
• Fixing HGVS conversion for indels (#452).

v0.30

Bugfix release.

jannovar-core

• Fixing interpretation of INFO/SVTYPE, urldecode and only use the first component after splitting at ":".
• Interpreting SV type annotation for SV2 more correctly.

v0.29

jannovar-core

• Putative impact of splice_region_variant has changed from MODERATE to LOW (see issue #439)
• Decreasing log verbosity in one location when building database.
• Fixing CDS region import in RefSeqParser
• Putative impact of splice_region_variant has changed from MODERATE to LOW (see issue #439)
• Fixing SV annotation using hg38/ucsc for transcripts without gene ID (see #444).
• Adding support for rn6 RefSeq transcripts.
  Adding allowNonCodingNm directive for data source INI file to disable check that RefSeq NM transcript has CDS.
• Adding versions to ENST accessions for ENSEMBL.

jannovar-vardbs

• Bugfix: TSVAnnotator did not use end given column.