Script for counting transcripts in RNA-Seq datasets with random n-mers in the adapter

[Koppology]

This will only serve a small niche of people doing RNA-seq, but figured I'd post it here anyway.

I wrote a script that essentially implements the counting described here:

http://www.pnas.org/content/early/2012/01/05/1118018109

If you have an n-mer of random nucleotides at the end of your RNA-seq adapter, you can use that n-mer to count the number of molecules of RNA. Basically, if you have.

This script takes a FASTQ file and collapses it based on the barcode. For example, if you have the reads

GAGAGAGAGGGGGGGGGG
GAGAGAGAGGGGGGGGGG
CACACACAGGGGGGGGGG

and the barcode is the first 8 nucleotides, it gets collapsed down to

GGGGGGGGGG
GGGGGGGGGG

since it is likely that the first two barcodes arose from PCRing the same molecule.

https://github.com/Koppology/count-transcripts

Requires: Python 2.7, numpy, and biopython.

Best,
David