GDC VEP Tool

Contains VEP plugins, utility scripts, and a dockerfile for running the VEP pipeline. See https://github.com/NCI-GDC/vep-cwl for more information about the GDC workflow associated with this.

VEP Plugins

We have only tested these plugins on the software dependencies listed in the Dockerfile:

• BioPerl 1.6.924
• VEP v84

All of these plugins require tabix to be installed and in your PATH which is often installed automatically (via htslib) during the VEP installation process.

Entrez

The GDC_entrez.pm plugin uses a pre-populated JSON file containing mappings from GENCODE to Entrez identifiers. The JSON file can be created using the vep-plugins/utils/make_ensembl_entrez_json.py script. The required input files for generating the JSON are:

• Gencode Entrez Gene IDs file: ftp://ftp.ebi.ac.uk/pub/databases/gencode/Gencode_human/release_22/gencode.v22.metadata.EntrezGene.gz
• NCBI human gene info file: ftp://ftp.ncbi.nih.gov/gene/DATA/GENE_INFO/Mammalia/Homo_sapiens.gene_info.gz

Note: The NCBI file is not versioned, so it may differ.

VEP Usage: --plugin GDC_entrez,<entrez_ensembl.json>

Evidence

The GDC_evidence.pm plugin uses the tabix-indexed VEP evidence VCF to add in the evidence values for dbSNP annotations.

VEP Usage: --plugin GDC_evidence,<variation.vcf.gz>