VarScan 2.4.2
VarScan is a platform-independent mutation caller for targeted, exome, and whole-genome resequencing data generated on Illumina, SOLiD, Life/PGM, Roche/454, and similar instruments. The newest version, VarScan 2, is written in Java, so it runs on most operating systems. http://dx.doi.org/10.1101/gr.129684.111 http://www.ncbi.nlm.nih.gov/pubmed/19542151
VarScan requires mpileup formatted input files, which are generally derived from BAM files. Since mpileup files can become humongous, the interim step of storing it is bypassed. Thus, in this wrapper one or multiple BAM/SAM files go in, get processed into a mpileup file and get directly linked to VarScan. The samtools package is not able to parallelize the mpileup generation which make it a very slow process. Other people were aware of this and have written a version that can do parallelization: https://github.com/mydatascience/parallel-mpileup
Consequently, when a BAM files gets processed by this wrapper, it's processed by parallel-mpileup before its send to VarScan.
Input formats
VarScan2 accepts sequencing alignments in the same, either SAM or BAM format (http://samtools.sourceforge.net/). The alignment files have to be linked to a reference genome by galaxy. This is indicated under every history item with e.g.: "database: hg19" for a link to hg19, or "database: ?" if the link is missing.
License
The tool wrapper has been written by Youri Hoogstrate from the Erasmus Medical Center (Rotterdam, Netherlands).