GATK4SCNA pipeline

This pipeline build for CPTAC2/3 projects to call somatic copy number alterations from whole exome sequencing (WXS) data. The pipeline implements panel of normals (PoN) approach.

• Version v0.1

  This version only used in MGI-Server (LSF).

Tutorial

• https://gatk.broadinstitute.org/hc/en-us/articles/360035531092#2
• https://gatk.broadinstitute.org/hc/en-us/articles/360035890011

Requirements

Install

GATK v4.1.9.0
bedtools v2.30.0
Python3

Set database

• Genome

  GRCh38.d1.vd1.fa & GRCh38.d1.vd1.dict

• Target

  targets.preprocessed.exome.interval_list

• Allelic data

  af-only-gnomad.hg38.common_biallelic.chr1-22XY.vcf

• Protein coding genes

  gencode.gene.info.v22.protein_coding.chr1-22.v2.tsv

[Note] Please refer to config/config.gatk4scna.mgi.ini

Input

* make table
  CaseID	NormalBam	TumorBam	Disease

Usage

[Note] Set config.ini location to gatk_somatic.cnv.mgi.sh

1. CollectFragmentCounts for Normal Bam
   `sh gatk_somatic.cnv.mgi.sh -p precall -t ./bam.catalog -o ./gatk4scna`
    or
    `sh gatk_somatic.cnv.mgi.sh -p precall -t ./bam.catalog -o ./gatk4scna -c ./config_file`
2. Make pool normal
   `sh gatk_somatic.cnv.mgi.sh -p pon -o ./gatk4scna`
   or
   `sh gatk_somatic.cnv.mgi.sh -p pon -o ./gatk4scna -c ./config_file`
    
3. Call tumor cnv based on pool normal
   `sh gatk_somatic.cnv.mgi.sh -p callcn -t bam.catalog -o ./gatk4scna`
    or
    `sh gatk_somatic.cnv.mgi.sh -p callcn -t bam.catalog -o ./gatk4scna -c ./config_file`
4. Call gene-level
   `sh gatk_somatic.cnv.mgi.sh -p geneLevel -t BRCA.paired.bam.catalog -o ./gatk4scna`
    or
    `sh gatk_somatic.cnv.mgi.sh -p geneLevel -t BRCA.paired.bam.catalog -o ./gatk4scna -c ./config_file`
5. Merge gene-level files to one file  
   `sh gatk_somatic.cnv.mgi.sh -p merge -o ./gatk4scna`
   or
   `sh gatk_somatic.cnv.mgi.sh -p merge -o ./gatk4scna -c ./config_file`

Contact

Hua Sun, hua.sun@wustl.edu