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 |