Rapid Autopsy project

Input: BAM files provided by TCAG for 3 normal, 3 diagnostic samples and 24 autopsy samples

Part 1: Somatic variant calling

https://github.com/kridel-lab/RAP_WGS/tree/master/Part_1_Somatic_variant_calling

• Ran Mutect2 which is optimized for identifying low frequency somatic variants
• Ran Manta to identify somatic structural variants for each sample
• Ran Strelka to idenify somatic SNVs for each sample
• Merge variants from Mutect2 and Strelka and apply further soft filters
  • minimum depth of 60
  • no indels
  • no chromosome X and Y
  • no population variants (Gnomad)
  • minimum VAF of 0.1
  • remove variants overlapping blacklisted regions (ENCODE)

Part 2: Somatic copy number calling

https://github.com/kridel-lab/RAP_WGS/tree/master/Part_2_Somatic_copy_number_calling

• Ran Sequenza to obtain overall copy number status, purity, ploidy and major minor CN status

Part 3: Analysis of variants, preliminary findings, basic summaries and visualizations

https://github.com/kridel-lab/RAP_WGS/tree/master/Part_3_Analysis_variants_preliminary_findings

• basic summary of variants per sample
• scripts for figures

Part 4: Phylogeny analysis

https://github.com/kridel-lab/RAP_WGS/tree/master/Part_4_Phylogeny_analysis

• Treeomics
• Pyclone-VI
• Pairtree

Part 5: ctDNA analysis

https://github.com/kridel-lab/RAP_WGS/tree/master/Part_5_ctDNA

• Ran ConsensusCruncher to process UMIs and generate BAM files
• Ran Mutect2 in tumour/normal mode to detect SNVs in plasma ctDNA (using WGS germline samples as controls)

Notes

• To easily load processed mutation and CNA data to work with on the cluster, use this script: https://github.com/kridel-lab/RAP_WGS/blob/master/config-file.R
• Full order of scripts in which they were run is here: https://github.com/kridel-lab/RAP_WGS/blob/master/order-scripts.sh