Biowulf at the NIH
RSS Feed


Strelka is an analysis package designed to detect somatic SNVs and small indels from the aligned sequencing reads of matched tumor-normal samples.

How to Use

There are multiple versions of Strelka available. An easy way of selecting the version is to use modules. To see the versions available, type

module avail strelka

To select a version, type

module load strelka/[ver]

where [ver] is the version of choice.

Strelka runs as a workflow. The workflow needs to be configured, then started using make. Example config files (*.ini) are available in $STRELKA_INSTALL_DIR/etc.

Once a config file has been created and/or edited, here is an example batch script for running the analysis:

# ----  this file is called strelka.qsub ---------
#PBS -N Strelka
#PBS -m be
#PBS -j oe

module load strelka

  --normal=/path/to/normal.bam \
  --tumor=/path/to/tumor.bam \
  --ref=/path/to/hg19.fa \
  --config=/path/to/config.ini \

# This example is run using 16 cores on the local host:

cd /path/to/myAnalysis
make -j 16

Several things must be changed in this script:

  • /path/to/normal.bam : Alignment (BAM) file for normal reads
  • /path/to/tumor.bam : Alignment (BAM) file for tumor reads
  • /path/to/hg19.fa : Reference (FASTA) file for the alignment
  • /path/to/config.ini : Configuration file for Strelka
  • /path/to/myAnalysis : Output directory
  • -j : Number of cores available on allocated node

This script assumes that the allocated node has at least 16 cores available. If this is not correct, change the value of the -j option, and alter the nodetype (below) to match.

Once the configuration file and batch script are in order, submit to the cluster like this:

qsub -l nodes=1:c16 strelka.qsub