j5kim/ensembl_vep repository overview

⁠IN THE HOST COMPUTER

⁠(Don't need to run anything in the Docker container. Do everything in the host computer.)

⁠Step 1. Pull the vep docker

docker pull j5kim/ensembl_vep

⁠Step 2. Download a big vep cache file (compressed file size = 4.7 GB) to a host computer

export LOCALCACHEDIR=/Users/johndoe/MyDockers/vep/prebuiltcache

cd ${LOCALCACHEDIR}

wget ftp://ftp.ensembl.org/pub/release-84/variation/VEP/homo_sapiens_vep_84_GRCh38.tar.gz

tar -zxvf homo_sapiens_vep_84_GRCh38.tar.gz

⁠Step 3. Download a real data for test running vep

cd ${LOCALCACHEDIR}

curl -O ftp://ftp-race.ncbi.nlm.nih.gov/giab/ftp/release/NA12878_HG001/latest/supplementaryfiles/NA12878_GIAB_highconf_CG-IllFB-IllGATKHC-Ion-Solid_ALLCHROM_v3.2_all.vcf.gz

gunzip NA12878_GIAB_highconf_CG-IllFB-IllGATKHC-Ion-Solid_ALLCHROM_v3.2_all.vcf.gz

cp NA12878_GIAB_highconf_CG-IllFB-IllGATKHC-Ion-Solid_ALLCHROM_v3.2_all.vcf GIAB_NA12878.vcf

⁠Step 4. Create a test run file in the HOST COMPUTER

cd ${LOCALCACHEDIR}

echo -e "cd /opt/vep/ensembl-tools-release-84/scripts/variant_effect_predictor" > testrun.sh

echo -e "perl variant_effect_predictor.pl --cache --dir /root/.vep -i /root/.vep/GIAB_NA12878.vcf -o /root/.vep/output_GIAB_NA12878.txt --chr 21 --coding_only" >> testrun.sh

⁠Step 5. Run docker with a real example, Genome In A Bottle (GIAB) NA12878 vcf file as an input

cd ${LOCALCACHEDIR}

docker run -i -t -m 4g -v ${LOCALCACHEDIR}:/root/.vep j5kim/ensembl_vep:v0.1 /bin/bash /root/.vep/testrun.sh

⁠Reference:

Zook et al, Nat Biotech, 2014 (http://www.nature.com/nbt/journal/v32/n3/full/nbt.2835.html⁠) Extensive sequencing of seven human genomes to characterize benchmark reference materials PMID = 24531798

https://docs.google.com/spreadsheets/d/1iL45zPit9-kVmk-9sDJEGMhxsUWf52n1nw_duTdNwcE/edit?pref=2&pli=1#gid=0⁠ https://sites.stanford.edu/abms/content/giab-reference-materials-and-data⁠