SomaticSeq is a flexible post-somatic-mutation-calling workflow for improved accuracy. It has incorporated multiple somatic mutation caller(s) to obtain a combined call set, and then it uses machine learning to distinguish true mutations from false positives from that call set. It has incorporated the following somatic mutation callers:
- Strelka, and
MetaCentrum provides following modules:
somaticseq-2.7.0(released: December 2017), NOTE: This module does not load any version of
R programming language- if you want to use this language,
you will need to load a module on your own
SomaticSeq is a "wrapper" which "puts together" a lot of other programs. Here you can find out, what exactly are SomaticSeq modules composed of.
SomaticSeq currently supports about 9 somatic mutation callers. In majority of cases, these callers are installed beside SomaticSeq (the only exception is
which is in totally different module):
Mutect2- part of the GATK 4 toolkit (which gets loaded automatically with the SomaticSeq module)
VarScan2- convenience wrapper script available, so after
module load somaticseq-2.7.0, you can execute
varscanand help will be printed.
The wrapper script works like this:
java -jar /software/somaticseq/2.7.0/VarScan/varscan.jar "$@"
JointSNVMix2- you can run this caller simply by executing the
jsm.pycommand. You can try
jsm.py -h trainor
jsm.py -h classifyto get help...
SomaticSniper- run any of the scripts directly, for example
VarDict- run any of the scripts directly, for example
MuSE <command> [options]
lofreq2_somatic.pyor any other caller's script
FindVariants.plor any other caller's script
strelka2or other caller's script
If you use SomaticSeq in your work, please cite the following: