Install Ensembl's VEP v106 with local cache for running offline

ensembl_vep_106_with_offline_cache.md

Ensembl's VEP (Variant Effect Predictor) is popular for how it picks a single effect per gene as detailed here, its CLIA-compliant HGVS variant format, and Sequence Ontology nomenclature for variant effects.

Instead of the official instructions, we will use mamba (conda, but faster) to install VEP and its dependencies. If you don't already have mamba, use these steps to download and install it into $HOME/mambaforge, then run a script that adds it to your $PATH:

curl -L https://github.com/conda-forge/miniforge/releases/download/4.12.0-0/Mambaforge-Linux-x86_64.sh -o /tmp/mambaforge.sh
sh /tmp/mambaforge.sh -bfp $HOME/mambaforge && rm -f mambaforge.sh
. $HOME/mambaforge/etc/profile.d/conda.sh

You can add the following to your ~/.bashrc file to add mamba and conda to your $PATH whenever you login:

if [ -f "$HOME/mambaforge/etc/profile.d/conda.sh" ]; then
    . $HOME/mambaforge/etc/profile.d/conda.sh
fi

Use mamba to create and activate a conda environment with VEP, its dependencies, and other related tools:

mamba create -n vep
conda activate vep
mamba install -y -c conda-forge -c bioconda -c defaults ensembl-vep==106.0 htslib==1.14 bcftools==1.14 samtools==1.14 ucsc-liftover==377

Download VEP's offline cache for GRCh38, and the reference FASTA:

mkdir -p $HOME/.vep/homo_sapiens/106_GRCh38/
rsync -avr --progress rsync://ftp.ensembl.org/ensembl/pub/release-106/variation/indexed_vep_cache/homo_sapiens_vep_106_GRCh38.tar.gz $HOME/.vep/
tar -zxf $HOME/.vep/homo_sapiens_vep_106_GRCh38.tar.gz -C $HOME/.vep/
rsync -avr --progress rsync://ftp.ensembl.org/ensembl/pub/release-106/fasta/homo_sapiens/dna_index/ $HOME/.vep/homo_sapiens/106_GRCh38/

(Optional) Download VEP's offline cache for GRCh37, and the reference FASTA which we must bgzip instead of gzip:

mkdir -p $HOME/.vep/homo_sapiens/106_GRCh37/
rsync -avr --progress rsync://ftp.ensembl.org/ensembl/pub/release-106/variation/indexed_vep_cache/homo_sapiens_vep_106_GRCh37.tar.gz $HOME/.vep/
tar -zxf $HOME/.vep/homo_sapiens_vep_106_GRCh37.tar.gz -C $HOME/.vep/
rsync -avr --progress rsync://ftp.ensembl.org/ensembl/pub/grch37/release-106/fasta/homo_sapiens/dna/Homo_sapiens.GRCh37.dna.toplevel.fa.gz $HOME/.vep/homo_sapiens/106_GRCh37/
gzip -d $HOME/.vep/homo_sapiens/106_GRCh37/Homo_sapiens.GRCh37.dna.toplevel.fa.gz
bgzip -i $HOME/.vep/homo_sapiens/106_GRCh37/Homo_sapiens.GRCh37.dna.toplevel.fa
samtools faidx $HOME/.vep/homo_sapiens/106_GRCh37/Homo_sapiens.GRCh37.dna.toplevel.fa.gz

Test running VEP in offline mode on a GRCh38 VCF:

curl -sLO https://raw.githubusercontent.com/Ensembl/ensembl-vep/release/106/examples/homo_sapiens_GRCh38.vcf
vep --species homo_sapiens --assembly GRCh38 --offline --no_progress --no_stats --sift b --ccds --uniprot --hgvs --symbol --numbers --domains --gene_phenotype --canonical --protein --biotype --tsl --pubmed --variant_class --shift_hgvs 1 --check_existing --total_length --allele_number --no_escape --xref_refseq --failed 1 --vcf --minimal --flag_pick_allele --pick_order canonical,tsl,biotype,rank,ccds,length --dir $HOME/.vep --fasta $HOME/.vep/homo_sapiens/106_GRCh38/Homo_sapiens.GRCh38.dna.toplevel.fa.gz --input_file homo_sapiens_GRCh38.vcf --output_file homo_sapiens_GRCh38.vep.vcf --polyphen b --af --af_1kg --af_esp --regulatory

Not sure why but mamba installed v107 of VEP instead so to use the cache I had to specify --cache_version 106