Releases: Illumina/Nirvana
Releases · Illumina/Nirvana
v3.18.1
Features
Issue key | Summary |
---|---|
NIR-1003 | Update ClinGen Dosage Sensitivity Map |
NIR-1002 | Update ClinGen Gene Disease Validity |
NIR-1001 | Update dbSNP to v155 |
NIR-1000 | Update OMIM to May 2022 |
NIR-999 | Update ClinVar to May 2022 |
NIR-969 | Add DECIPHER SVs to Nirvana |
NIR-926 | Remove Variant recomposition |
Bugs Fixed
Issue key | Summary |
---|---|
NIR-1038 | Nirvana displays confusing error message if idx files for gsa files are absent |
NIR-1014 | Downloader is not using new SA manifest architecture |
NIR-996 | Nirvana returns duplicate VCVs in ClinVar |
NIR-976 | Jasix crashes if range is out of the range in the file |
Unresolved Bugs
Issue key | Summary |
---|---|
NIR-919 | Nirvana does not handle protein sequences that are partially aligned to the genome |
NIR-918 | Nirvana does not handle ribosomal frameshifts properly during translation |
NIR-917 | Nirvana does not truncate deletions that start within a deletion RNA-edit (HGVS c.) |
NIR-916 | Nirvana does not detect HGVS c. duplications when inserting a base into a deletion RNA-edit |
NIR-915 | Nirvana does not convert SNVs that land in a deletion RNA-edit as an insertion (HGVS c.) |
NIR-909 | Nirvana Lambda throws Nirvana error if input VCF is not BGZipped |
NIR-701 | Nirvana should simplify HGVS p. notation for deletions that disrupt the initiation codon |
NIR-702 | Nirvana should stop after inserted terminal codon in HGVS p. |
NIR-703 | Nirvana incorrectly handles deletions across the terminal codon in HGVS p. |
NIR-667 | Nirvana is unable to count the # of AA until new stop codon (HGVS p. frameshift) |
NIR-666 | Nirvana creates frameshift HGVS p. notation when extension is expected |
NIR-665 | Nirvana uses incorrect HGVS p. notation for disrupted translation initiation codons |
NIR-602 | SAUtils CustomVar won't allow missing values for Score columns |
NIR-504 | SingleAnnotation behaves differently in Lambda vs REST API |
NIR-220 | Incorrect HGVS p. results because Nirvana does not use the transcript sequence |
NIR-219 | Nirvana reports incorrect HGVS p. for non-AUG start codon variants |
NIR-215 | Nirvana incorrectly determines gnomAD low complexity regions at translocation breakend boundaries |
NIR-86 | Nirvana incorrectly handles insertions where RNA-edit insertions already occur |
NIR-56 | CDS position goes beyond the CDS sequence length (NM_001300952.1) |
v3.18.0
Features
Issue key | Summary |
---|---|
NIR-987 | Support Custom Sample Format fields |
NIR-927 | Update Nirvana to use net6.0 |
NIR-826 | Update ClinGen gene disease validity to Feb 14, 2022 version |
NIR-825 | Update ClinGen dosage sensitivity to Feb 14, 2022 version |
NIR-824 | Update OMIM to Dec 02, 2021 version |
NIR-823 | Update ClinVar to Dec 2021 |
NIR-783 | Support Custom INFO fields |
NIR-779 | Add support for DANN scores |
NIR-778 | Add support for GERP++ conservation scores |
NIR-775 | Add support for GME Variome |
NIR-773 | Add support for gnomAD 2.1 SV |
Bugs Fixed
Issue key | Summary |
---|---|
NIR-975 | Nirvana JSON uses inconsistent decimal identifiers |
NIR-913 | Nirvana lambda signature expires if Annotation lambda takes more than 5 mins |
NIR-912 | Nirvana missing triplosensitivity in annotation |
NIR-874 | GeneAnnotationLambda always uses 'latest' manifest |
NIR-810 | Custom Annotation should provide input line when an error is encountered |
NIR-759 | Nirvana lambda throws Nirvana error when signature expires |
NIR-582 | Jasix incorrectly closes the JSON header when combining query with header |
NIR-479 | Single Annotation lambda should handle incorrect alt alleles more robustly |
Unresolved Bugs
Issue key | Summary |
---|---|
NIR-996 | Nirvana returns duplicate VCVs in ClinVar |
NIR-976 | Jasix crashes if range is out of the range in the file |
NIR-919 | Nirvana does not handle protein sequences that are partially aligned to the genome |
NIR-918 | Nirvana does not handle ribosomal frameshifts properly during translation |
NIR-917 | Nirvana does not truncate deletions that start within a deletion RNA-edit (HGVS c.) |
NIR-916 | Nirvana does not detect HGVS c. duplications when inserting a base into a deletion RNA-edit |
NIR-915 | Nirvana does not convert SNVs that land in a deletion RNA-edit as an insertion (HGVS c.) |
NIR-909 | Nirvana Lambda throws Nirvana error if input VCF is not BGZipped |
NIR-701 | Nirvana should simplify HGVS p. notation for deletions that disrupt the initiation codon |
NIR-702 | Nirvana should stop after inserted terminal codon in HGVS p. |
NIR-703 | Nirvana incorrectly handles deletions across the terminal codon in HGVS p. |
NIR-667 | Nirvana is unable to count the # of AA until new stop codon (HGVS p. frameshift) |
NIR-666 | Nirvana creates frameshift HGVS p. notation when extension is expected |
NIR-665 | Nirvana uses incorrect HGVS p. notation for disrupted translation initiation codons |
NIR-602 | SAUtils CustomVar won't allow missing values for Score columns |
NIR-504 | SingleAnnotation behaves differently in Lambda vs REST API |
NIR-220 | Incorrect HGVS p. results because Nirvana does not use the transcript sequence |
NIR-219 | Nirvana reports incorrect HGVS p. for non-AUG start codon variants |
NIR-215 | Nirvana incorrectly determines gnomAD low complexity regions at translocation breakend boundaries |
NIR-86 | Nirvana incorrectly handles insertions where RNA-edit insertions already occur |
NIR-56 | CDS position goes beyond the CDS sequence length (NM_001300952.1) |
3.2.6
This is the latest release in our Nirvana 3.2.x branch series used by the TruSight Oncology team at Illumina. This release uses the latest and greatest transcripts and genes directly from RefSeq.
Running Nirvana
Nirvana 3.2.6 has some difference compared with Nirvana 3.2.5:
- the cache data now comes directly from RefSeq rather than from VEP. As such there are slight differences. For example, RefSeq uses official Sequence Ontology terms for the transcript biotypes rather than the related Ensembl terminology.
- Phantom variant recomposition has been disabled for the time being
- SIFT and PolyPhen are currently not available in the normal cache, but they are available for the transcripts in our CDx cache
- the Nirvana command-line now asks for a cache directory rather than a cache path prefix
A typical Nirvana 3.2.6 command-line might look like this:
dotnet bin/Nirvana.dll -i NA12878.vcf.gz -c Data/Cache/GRCh37 -r Data/References/Homo_sapiens.GRCh37.Nirvana.dat --sd Data/SupplementaryAnnotation/GRCh37 -o NA12878
Installation
Download the files
mkdir Nirvana-3.2.6
cd Nirvana-3.2.6
wget https://nirvana-prerelease.s3.us-west-2.amazonaws.com/Reference-GRCh37.zip
wget https://github.com/Illumina/Nirvana/releases/download/v3.2.6/Nirvana-3.2.6-dotnet-6.0.zip
wget https://nirvana-prerelease.s3.us-west-2.amazonaws.com/Cache-GRCh37-RefSeq.zip
wget https://nirvana-prerelease.s3.us-west-2.amazonaws.com/Cache-GRCh37-CDx.zip
wget https://nirvana-prerelease.s3.us-west-2.amazonaws.com/Cache-HGNC-GeneSymbols.zip
wget https://nirvana-prerelease.s3.us-west-2.amazonaws.com/Cache-TSO500-GeneSymbols.zip
wget https://nirvana-prerelease.s3.us-west-2.amazonaws.com/Sift-CDx-VEP91.zip
wget https://nirvana-prerelease.s3.us-west-2.amazonaws.com/PolyPhen-CDx-VEP91.zip
Unpack Nirvana
mkdir -p bin Data/Cache/GRCh37 Data/References Data/SupplementaryAnnotation/GRCh37
cd bin
unzip ../Nirvana-3.2.6-dotnet-6.0.zip
Unpack the data files
cd ../Data/Cache/GRCh37
unzip ../../../Cache-GRCh37-RefSeq.zip
unzip ../../../Cache-GRCh37-CDx.zip
unzip ../../../Cache-TSO500-GeneSymbols.zip
cd ../../References
unzip ../../Reference-GRCh37.zip
cd ../SupplementaryAnnotation/GRCh37
unzip ../../../Sift-CDx-VEP91.zip
unzip ../../../PolyPhen-CDx-VEP91.zip
cd ../../..
In the example above, we're unpacking the gene symbols (Cache-TSO500-GeneSymbols.zip
) that were used in Nirvana 3.2.5. If you want the latest gene symbols from HGNC as of 2022-02-21, unpack the Cache-HGNC-GeneSymbols.zip
file instead.
Checking the installation
dotnet bin/Nirvana.dll
You should see the following:
---------------------------------------------------------------------------
Nirvana (c) 2022 Illumina, Inc.
Stromberg, Roy, Siddiqui, Platzer, et al. 3.2.6
---------------------------------------------------------------------------
USAGE: dotnet Nirvana.dll -i <vcf path> -c <cache dir> --sd <sa dir> -r <ref path> -o <base output filename>
Annotates a set of variants
OPTIONS:
--cache, -c <dir> input cache dir
--in, -i <path> input VCF path
--gvcf enables genome vcf output
--vcf enables vcf output
--out, -o <file path> output file path
--ref, -r <path> input compressed reference sequence path
--sd <directory> input supplementary annotation directory
--force-mt forces to annotate mitochondrial variants
--help, -h displays the help menu
--version, -v displays the version
Cache version: 1, Supplementary annotation version: 50, Reference version: 7
If this doesn't work, you might need to install either the .NET 6.0 runtime or the .NET 6.0 SDK. However, if you're only running Nirvana, you only need the runtime.
Features
Issue key | Summary |
---|---|
NIR-920 | Add support for MT transcripts |
NIR-879 | Add gene symbol override to Nirvana |
NIR-876 | Add support for discovering amino acid edits (non-canonical start codons, etc.) |
NIR-872 | Add support for RefSeq regulatory regions |
NIR-798 | Add support for multiple caches belonging to the same genome assembly |
NIR-438 | Create RefSeq Cache v28 files |
Bugs fixed
Issue key | Summary |
---|---|
NIR-1036 | Resolve consequence issues that were identified in VEP 105 comparison |
NIR-703 | Nirvana incorrectly handles deletions across the terminal codon in HGVS p. |
NIR-56 | CDS position goes beyond the CDS sequence length (NM_001300952.1) |
Unresolved bugs
Issue key | Summary | Fix Version |
---|---|---|
NIR-919 | Nirvana does not handle protein sequences that are partially aligned to the genome | |
NIR-918 | Nirvana does not handle ribosomal frameshifts properly during translation | |
NIR-721 | Nirvana does not properly preload <NON_REF> variants | 3.9.0 |
NIR-714 | Nirvana should trim and LS variants during TopMed lift-over using dbSNP IDs | 3.9.0 |
NIR-719 | Nirvana does not remove missing values in 1KG SV IDs | 3.8.0 |
NIR-722 | Nirvana does not throw error when annotating unsorted VCF file | 3.6.0 |
NIR-718 | Nirvana reports zero allele frequencies for 1KG SV supplementary annotations | 3.6.0 |
NIR-717 | Nirvana does not annotate properly when variants fall into transcript gaps | 3.5.0 |
NIR-713 | Nirvana cannot handle OMIM phenotype 617956 (β-glucopyranoside tasting) | 3.5.0 |
NIR-720 | Nirvana creates empty custom annotations | 3.3.0 |
NIR-716 | Nirvana should assign copy_number_loss to <DEL> with SVTYPE=CNV | 3.3.0 |
NIR-715 | Nirvana provides extra PhyloP scores for some genomic positions | 3.3.0 |
NIR-712 | Nirvana is missing dbSNP annotations for mitochondrial variants | 3.3.0 |
NIR-701 | Nirvana should simplify HGVS p. notation for deletions that disrupt the initiation codon | |
NIR-86 | Nirvana should use the transcript sequence for HGVS c. instead of reference |
3.17
Features
Issue key | Summary |
---|---|
NIR-678 | Update OMIM to the August 2021 version |
NIR-674 | Update ClinVar to the August 2021 version |
NIR-671 | Optimize memory for JSON outputs |
Bugs fixed
Issue key | Summary |
---|---|
NIR-724 | Buffer overrun in BgzipTextWriter |
NIR-663 | Nirvana λ is unable to handle empty VCF files |
NIR-549 | JWT validator doesn't work for api regions |
NIR-212 | Annotation lambdas run out of disk space when large amounts of CA are submitted |
Unresolved bugs
Issue key | Summary |
---|---|
NIR-759 | Nirvana lambda throws Nirvana error when signature expires |
NIR-703 | Nirvana incorrectly handles deletions across the terminal codon in HGVS p. |
NIR-702 | Nirvana should stop after inserted terminal codon in HGVS p. |
NIR-701 | Nirvana should simplify HGVS p. notation for deletions that disrupt the initiation codon |
NIR-667 | Nirvana is unable to count the # of AA until new stop codon (HGVS p. frameshift) |
NIR-666 | Nirvana creates frameshift HGVS p. notation when extension is expected |
NIR-665 | Nirvana uses incorrect HGVS p. notation for disrupted translation initiation codons |
NIR-602 | SAUtils CustomVar won't allow missing values for Score columns |
NIR-582 | Jasix incorrectly closes the JSON header when combining query with header |
NIR-504 | SingleAnnotation behaves differently in Lambda vs REST API |
NIR-479 | Single Annotation lambda should handle incorrect alt alleles more robustly |
NIR-220 | Incorrect HGVS p. results because Nirvana does not use the transcript sequence |
NIR-219 | Nirvana reports incorrect HGVS p. for non-AUG start codon variants |
NIR-215 | Nirvana incorrectly determines gnomAD low complexity regions at translocation breakend boundaries |
NIR-86 | Nirvana incorrectly handles insertions where RNA-edit insertions already occur |
NIR-56 | CDS position goes beyond the CDS sequence length (NM_001300952.1) |
3.2.5.1
Features
Issue key | Summary |
---|---|
NIR-636 | Fix transcripts with translational readthrough (AA difference) |
NIR-631 | Add support for non-AUG start codons |
NIR-627 | Fix transcripts with cDNA frameshifts |
NIR-625 | Fix transcripts with cDNA substitutions |
NIR-624 | Fix transcripts with missing poly A tails in cDNA |
NIR-456 | Remove reference bases from HGVS notation |
Bugs fixed
Issue key | Summary |
---|---|
NIR-702 | Nirvana should stop after inserted terminal codon in HGVS p. |
NIR-667 | Nirvana is unable to count the # of AA until new stop codon (HGVS p. frameshift) |
NIR-666 | Nirvana creates frameshift HGVS p. notation when extension is expected |
NIR-665 | Nirvana uses incorrect HGVS p. notation for disrupted translation initiation codons |
NIR-484 | Incorrect gene models were found for NM_022148.2 |
NIR-220 | Incorrect HGVS p. results because Nirvana does not use the transcript sequence |
NIR-219 | Nirvana reports incorrect HGVS p. for non-AUG start codon variants |
NIR-57 | Nirvana produces bizarre HGVS p. notation (NP_001137434.1:p.(Val701_?_fsTer?)) |
NIR-55 | Nirvana produces invalid CDS positions |
Unresolved bugs
Issue key | Summary | Fix Version |
---|---|---|
NIR-721 | Nirvana does not properly preload <NON_REF> variants | 3.9.0 |
NIR-714 | Nirvana should trim and LS variants during TopMed lift-over using dbSNP IDs | 3.9.0 |
NIR-719 | Nirvana does not remove missing values in 1KG SV IDs | 3.8.0 |
NIR-722 | Nirvana does not throw error when annotating unsorted VCF file | 3.6.0 |
NIR-718 | Nirvana reports zero allele frequencies for 1KG SV supplementary annotations | 3.6.0 |
NIR-717 | Nirvana does not annotate properly when variants fall into transcript gaps | 3.5.0 |
NIR-713 | Nirvana cannot handle OMIM phenotype 617956 (β-glucopyranoside tasting) | 3.5.0 |
NIR-720 | Nirvana creates empty custom annotations | 3.3.0 |
NIR-716 | Nirvana should assign copy_number_loss to <DEL> with SVTYPE=CNV | 3.3.0 |
NIR-715 | Nirvana provides extra PhyloP scores for some genomic positions | 3.3.0 |
NIR-712 | Nirvana is missing dbSNP annotations for mitochondrial variants | 3.3.0 |
NIR-703 | Nirvana incorrectly handles deletions across the terminal codon in HGVS p. | |
NIR-701 | Nirvana should simplify HGVS p. notation for deletions that disrupt the initiation codon | |
NIR-86 | Nirvana should use the transcript sequence for HGVS c. instead of reference | |
NIR-56 | CDS position goes beyond the CDS sequence length (NM_001300952.1) |
3.16.1
Features
Issue key | Summary |
---|---|
NIR-630 | Increased gzip compression level |
NIR-651 | Renamed TCGA & TumorFusions data sources |
Bugs fixed
Unresolved bugs
Issue key | Summary |
---|---|
NIR-602 | SAUtils CustomVar won't allow missing values for Score columns |
NIR-582 | Jasix incorrectly closes the JSON header when combining query with header |
NIR-504 | SingleAnnotation behaves differently in Lambda vs REST API |
NIR-220 | Nirvana doesn't use the transcript sequence to report the reference amino acids in HGVS p. notations |
NIR-219 | Nirvana reports ambiguous protein level annotations for CUG start codon variants |
NIR-218 | Duplicates may be marked as insertions in HGVS |
NIR-217 | Nirvana produces HGVS insertions notations that are more than one bp apart |
NIR-215 | Nirvana incorrectly determines gnomAD low complexity regions at translocation breakend boundaries |
NIR-212 | Annotation lambdas run out of disk space when large amounts of CA are submitted |
NIR-91 | Nirvana should fully right align duplications in HGVS c. notation |
NIR-87 | Nirvana should detect duplications in HGVS by looking at the transcript sequence. |
NIR-86 | Nirvana incorrectly handles insertions where RNA-edit insertions already occur |
NIR-75 | Nirvana doesn't handle 3' UTR intron HGVS c. notation properly |
NIR-57 | Nirvana produces bizarre HGVS p. notation (NP_001137434.1:p.(Val701_?_fsTer?)) |
NIR-56 | CDS position goes beyond the CDS sequence length (NM_001300952.1) |
NIR-55 | Nirvana produces a strange CDS position range for NM_001346067.1 |
3.15
Features
Issue key | Summary |
---|---|
NIR-574 | Update OMIM |
NIR-569 | Update ClinVar |
NIR-540 | Add command-line arguments to the Downloader for changing S3 URLs and manifest filenames |
NIR-520 | Add support for identifying in-frame gene fusions |
NIR-495 | Port bidirectional allele trimmer fix for breakend locations when using legacy VID mode |
NIR-494 | Port legacy VID fixes from 3.6.x |
NIR-480 | Add support for INV3/INV5 tags with respect to gene fusions |
NIR-477 | Add support for FusionCatcher data sources |
NIR-474 | Add support for COSMIC gene fusions |
NIR-459 | Update HGVS g. notation for MT variants to use m. notation |
NIR-324 | Allow masked reference bases in PAR regions (chrY) |
NIR-257 | Update Nirvana to use gnomAD 3.1 for GRCh38 |
NIR-250 | Update the BlockCompression library |
NIR-193 | Add support for detecting out of disk space issues in Downloader |
Bugs fixed
Issue key | Summary |
---|---|
NIR-548 | Nirvana incorrectly recomposes an deletion and assigns a negative genotype |
NIR-398 | Mito Heteroplasmy not using trimmed and left-aligned alt alleles for annotation |
NIR-214 | Nirvana doesn't support regular gzip VCF file |
NIR-213 | Downloader unintentionally removes genomic reference file |
Unresolved bugs
Issue key | Summary |
---|---|
NIR-602 | SAUtils CustomVar won't allow missing values for Score columns |
NIR-582 | Jasix incorrectly closes the JSON header when combining query with header |
NIR-504 | SingleAnnotation behaves differently in Lambda vs REST API |
NIR-220 | Nirvana doesn't use the transcript sequence to report the reference amino acids in HGVS p. notations |
NIR-219 | Nirvana reports ambiguous protein level annotations for CUG start codon variants |
NIR-218 | Duplicates may be marked as insertions in HGVS |
NIR-217 | Nirvana produces HGVS insertions notations that are more than one bp apart |
NIR-215 | Nirvana incorrectly determines gnomAD low complexity regions at translocation breakend boundaries |
NIR-212 | Annotation lambdas run out of disk space when large amounts of CA are submitted |
NIR-91 | Nirvana should fully right align duplications in HGVS c. notation |
NIR-87 | Nirvana should detect duplications in HGVS by looking at the transcript sequence. |
NIR-86 | Nirvana incorrectly handles insertions where RNA-edit insertions already occur |
NIR-75 | Nirvana doesn't handle 3' UTR intron HGVS c. notation properly |
NIR-57 | Nirvana produces bizarre HGVS p. notation (NP_001137434.1:p.(Val701_?_fsTer?)) |
NIR-56 | CDS position goes beyond the CDS sequence length (NM_001300952.1) |
NIR-55 | Nirvana produces a strange CDS position range for NM_001346067.1 |
3.14
Features
Issue key | Summary |
---|---|
NIR-210 | Cap VRF to 0.999 |
Bugs fixed
Unresolved bugs
Issue key | Summary |
---|---|
NIR-220 | Nirvana doesn't use the transcript sequence to report the reference amino acids in HGVS p. notations |
NIR-219 | Nirvana reports ambiguous protein level annotations for CUG start codon variants |
NIR-218 | Duplicates may be marked as insertions in HGVS |
NIR-217 | Nirvana produces HGVS insertions notations that are more than one bp apart |
NIR-216 | Nirvana reports extra ClinVar SNVs where there is no alternate allele in XML source file |
NIR-215 | Nirvana incorrectly determines gnomAD low complexity regions at translocation breakend boundaries |
NIR-214 | Nirvana doesn't support regular gzip VCF file |
NIR-213 | Downloader unintentionally removes genomic reference file |
NIR-212 | Annotation lambdas run out of disk space when large amounts of CA are submitted |
NIR-91 | Nirvana should fully right align duplications in HGVS c. notation |
NIR-87 | Nirvana should detect duplications in HGVS by looking at the transcript sequence. |
NIR-86 | Nirvana incorrectly handles insertions where RNA-edit insertions already occur |
NIR-75 | Nirvana doesn't handle 3' UTR intron HGVS c. notation properly |
NIR-57 | Nirvana produces bizarre HGVS p. notation (NP_001137434.1:p.(Val701_?_fsTer?)) |
NIR-56 | CDS position goes beyond the CDS sequence length (NM_001300952.1) |
NIR-55 | Nirvana produces a strange CDS position range for NM_001346067.1 |
3.13
- Major increase in annotation accuracy for RefSeq transcripts. 59% reduction in HGVS issues.
- Better support for reference variants when producing HGVS g. notation
3.12
- Updated ClinVar & OMIM
- Improved error handling when using the gene custom annotation files
- Added support for gnomAD 3.0 (but not active due to file size limitations in our CDN)
- Added support for handling custom annotation genomic regions that also include symbolic alleles