msisensor

Microsatelite Instability (MSI) detection using msisensor-pro

Overview

Dependencies

• msisensorpro 1.2.0

Usage

Cromwell

java -jar cromwell.jar run msisensor.wdl --inputs inputs.json

Inputs

Required workflow parameters:

Parameter	Value	Description
normalbam	File	normal input .bam file
normalbai	File	normal input .bai file
tumorbam	File	tumor input .bam file
tumorbai	File	tumor input .bai file
boostrapIt	Boolean	Enable bootstrapping
reference	String	reference genome of input sample

Optional workflow parameters:

Parameter	Value	Default	Description
outputFileNamePrefix	String	basename("~{tumorbam}",".bam")	Base name

Optional task parameters:

Parameter	Value	Default	Description
runMSIsensor.outputFileNamePrefix	String	basename("~{tumorbam}",".bam")	Base name
runMSIsensor.difficultRegions	String?	None	Path to .bed of difficult regions to exclude
runMSIsensor.jobMemory	Int	64	Memory allocated for this job (GB)
runMSIsensor.threads	Int	4	Requested CPU threads
runMSIsensor.timeout	Int	10	Hours before task timeout
bootstrapMSIsensor.boots	Int	100	number of bootstraps
bootstrapMSIsensor.loci	Int	500	number of loci to include in each bootstrap
bootstrapMSIsensor.outputFileNamePrefix	String	basename("~{tumorbam}",".bam")	Base name
bootstrapMSIsensor.jobMemory	Int	64	Memory allocated for this job (GB)
bootstrapMSIsensor.threads	Int	4	Requested CPU threads
bootstrapMSIsensor.timeout	Int	10	Hours before task timeout
bootstrapMSIsensor.difficultRegions	String?	None	bed file of regions to avoid, if necessary

Outputs

Output	Type	Description	Labels
msiGermline	File	A poorly documented output, ostensibly germline-specific metrics for MSI sites	vidarr_label: msiGermline
msiSomatic	File	A poorly documented output, ostensibly somatic-specific metrics for MSI sites	vidarr_label: msiSomatic
msiFinalOutput	File	Final msisensor call as .tsv, last column is msi score	vidarr_label: msiFinalOutput
msibooted	File?	msisensor calls bootstrapped	vidarr_label: msibooted

Commands

This section lists command(s) run by msisensor workflow

• Running msisensor

Microsatelite Instability (MSI) detection using msisensor-pro. MSIsensor-pro is an updated version of msisensor. MSIsensor-pro evaluates Microsatellite Instability (MSI) for cancer patients with NGS data. It accepts the whole genome sequencing, whole exome sequencing and target region (panel) sequencing data as input. MSIsensor-pro introduces a multinomial distribution model to quantify polymerase slippages for each tumor sample and a discriminative sites selection method to enable MSI detection without matched normal samples.

Run msisensor-pro (the main step)

  set -euo pipefail
 
  msisensor-pro msi
    -d MSIFILE 
    -e difficultRegions (Optional)
    -n NORMAL_BAM 
    -t TUMOR_BAM
    -o BASENAME.msi 
 

Optional bootstrapping with the same inputs:

  set -euo pipefail
 
  for boot in {1..~{boots}}
  do
  shuf -n ~{loci} ~{msifile} > rep.list
 
  sort -k1,1 -k2,2n rep.list > rep.list.sorted
 
  msisensor-pro msi 
   -d rep.list.sorted 
   -e difficultRegions (Optional)
   -n NORMAL_BAM
   -t TUMOR_BAM
   -o BASENAME.msi
 
  awk -v boot="${boot}" '$1 !~ "Total_Number_of_Sites" {print boot"\t"$1"\t"$2"\t"$3}' ~{basename}.msi >> ~{basename}.msi.booted
  done
 

Support

For support, please file an issue on the Github project or send an email to [email protected] .

Generated with generate-markdown-readme (https://github.com/oicr-gsi/gsi-wdl-tools/)