GMAP vs GSNAP

[Artur Jaroszewicz]

Hello all,

I was wondering if someone could enlighten me as to the difference between GMAP and GSNAP, their advantages and disadvantages, and usage difference. I'm running RNA-seq data with 50 bp single-end reads, and am mostly interested in gene expression, and soon isoform expression. I've searched all the documentation, and all I've found is that GSNAP is for short reads, which to me sounds quite ambiguous.

Also, as a side note, I was wondering what the output of GMAP is... I know GSNAP requires a redirection to another file; does GMAP write to stdout or to a new file? I'm interested in the SAM output, as my current pipelines are all set up for that format.

Thank you greatly in advance,
Artur

[darked89]

GSNAP is what you can use for spliced mapping of Illumina 50bp single-end reads. The default output is SAM. With 50bp you will have some problems with reads spanning splice junctions, so you may be better of if you map your reads to transcriptome in an unspliced mode first, like in tophat 1.4.0.

BTW, gsnap --help gives you more options.

GMAP is for ESTs/cDNAs mapping, and a sensible compromise between speed and sensitivity, i.e as compared to exonerate (way slower).