Workflows to process raw sequence into VCF

• Forked over from https://github.com/davemcg/biowulf2-bin

For NISC exomes/WGS

• Refer to readme.md in NISC_workflow/

Overview

1. (Re)Align with bwa-mem against 1000G GRCh37 with decoy

• NGS_generic_workflow/run_bwa-mem_hg37d5_fromBam.sh (if starting from a bam)
• NGS_generic_workflow/run_bwa-mem_hg37d5.sh (if starting from fastq)

2. Call GVCF with GATK (v3.5-0 right now)

• src/process_and_callGVCF.sh

3. Filter GVCF to a master VCF

• cp src/Snakefile_gvcf_to_vcf_example_config.yaml to your working directory
• mkdir GVCFs
• run ln -s /path/to/your/gvcfs/*.raw.g.vcf GVCFs
  • alternatively, just copy them, but that's not a great use of disk space
• edit Snakefile_gvcf_to_vcf_example_config.yaml to give ped path and whether you want VQSR or hard filtering (yaml file has some explanatory comments)
• run the Snakefile: sbatch --time=24:00:00 ~/git/NGS_genotype_calling/GVCF_to_VCF_snakemake.wrapper.sh Snakefile_gvcf_to_vcf_example_config.yaml

4. Annotate variants

• New repository: https://github.com/davemcg/variant_prioritization