SRA download to Google Cloud Storage

sra_download.yaml

name: sra_download
description: use Google Pipeline API to download an SRA run, reformat it as unaligned BAM, and upload it to Google Cloud Storage.  Run it like this: gcloud alpha genomics pipelines run --inputs SAMPLE=XXXXX --inputs RUN=XXXXX --outputs OUTPUT_FILE=gs://XXXXX --pipeline-file=sra_download.yaml
resources:
  #increase boot disk from 10GB to 50GB to accomodate intermediate files
  bootDiskSizeGb: 50
  #specify multiple zones so this pipeline will run in parallel
  zones:
  - us-west1-a
  - us-west1-b
  - us-east1-b
  - us-east1-c
  - us-east1-d
  - us-central1-a
  - us-central1-b
  - us-central1-c
  - us-central1-f
inputParameters:
- name: SAMPLE
- name: RUN
outputParameters:
- name: OUTPUT_FILE
  defaultValue: gs://allenday-dev/oryza/primary/${SAMPLE}/${RUN}.bam
  localCopy:
    disk: boot
    path: /tmp/output.bam
docker:
  imageName: index.docker.io/allenday/bfx-seq
  cmd: "cd /tmp && /opt/sratoolkit/bin/fastq-dump --split-files -F -O . ${RUN} && picard-tools FastqToSam F1=${RUN}_1.fastq F2=${RUN}_2.fastq O=${RUN}.unsort.bam SAMPLE_NAME='${SAMPLE}' READ_GROUP_NAME='${RUN}' && samtools sort ${RUN}.unsort.bam ${RUN}.sort && picard-tools MarkDuplicates I=${RUN}.sort.bam O=output.bam M=${RUN}.metrics"