Looking to find number overlapping sequences between fastq files

[jme]

Hi all, just starting to analyse NGS data, beginning to realise I should have been here asking for help from the beginning!

I have a number of fastq files (produced by concatenating both directions of a paired-end Illumina run), from a variety of PCRs from different samples. I'd like to be able to compare the files, to see how many sequences in one file are shared with say one or more other files.

Can anyone think of a sensible way of going about this? Bioinformatics is really not my strong suit, and the only ways I can think of would probably never run on my machine.

[kopi-o]

I would start by looking at something like vmatch, unless you are looking at the simplest case where you just want to identify exactly identical sequences. In this case, you should be able to do it with regular UNIX command-line tools like "sort" (after having pulled out just the sequence lines in the FASTQ files) and "comm"or "join".

[jme]

Vmatch looks great, thanks - I'll get in touch with the guys there about a copy.

In the meantime, 'comm' looks like it could do a wonderful job! But ideally I'd like to allow for mismatches, so something like vmatch looks much more up my street.

Cheers!