dajunluo/deepvariant Docker 镜像 - 轩辕镜像 | Docker 镜像高效稳定拉取服务

Run Google's DeepVariant locally

• No need for Google Cloud Platform.
• A standard model (Inception-V3) for WGS data is already included.
• Does not support GPU so far.
• Current version r0.8.0.

Demo (ucsc.hg19.chr20)

shell
$ export OUTPUT_DIR=/path/to/output/folder/
$ docker run -v "${OUTPUT_DIR}":/opt/deepvariant/output dajunluo/deepvariant python run_deepvariant.py \
--model_type=WGS \
--ref=../../quickstart-testdata/ucsc.hg19.chr20.unittest.fasta \
--reads=../../quickstart-testdata/NA***_S1.chr20.10_10p1mb.bam \
--regions "chr20:10,000,000-10,010,000" \
--output_vcf=../output/output.vcf.gz \
--output_gvcf=../output/output.g.vcf.gz
$ ls $OUTPUT_DIR

The output .vcf should be in OUTPUT_DIR now.

Using Your Own Data

Start the container

shell
$ export DATA_DIR='abosulute/path/to/your/bam/and/fasta'
$ OUTPUT_DIR=/path/to/output/folder/
$ docker run -td --name deepvariant -v "${DATA_DIR}":/opt/data -v "${OUTPUT_DIR}":/opt/deepvariant/output dajunluo/deepvariant

Enter the image

shell
$ docker exec -it deepvariant /bin/bash

Inside the docker image

shell
$python run_deepvariant.py --model_type=WGS \
--ref=../../data/"${REFERENCE_FILE}" \
--reads=../../data/"${BAM_FILE}" \
--regions "chr20:10,000,000-10,010,000" \
--output_vcf=../output/output.vcf.gz \
--output_gvcf=../output/output.g.vcf.gz

The output .vcf should be in OUTPUT_DIR now.

Also, there is an official image provided by Google at [***]

contact: ***