Use jellyfish with paired end sequencing data

Question

[db]

I am wondering how exactly we should run jellyfish to count k-mers for paired end sequencing data. Is it enough to concatenate the forward and reverse read (using cat) or do we need to merge the reads (using something like PEAR)?

Answer 1

[Brian Bushnell]

Concatenating the files works fine (I think you can just give it multiple files, but the syntax in the help message is a little confusing). Merging the reads is also a possibility, but you don't really need to do so in this case, and note that it would change the counts and yield even more files (merged, r1 unmerged, r2 unmerged).

[db]

Thank you Brian. Are you saying it is ok to concatenate based on your experience (not that I don't trust you)? It's just that I could not find any documentation on how to deal with paired end data.

[Brian Bushnell]

For pure kmer counting, read pairing does not matter, and concatenated fastq files are still valid fastq files. I have never actually concatenated two files and then run Jellyfish on the result, but I have done that with various other programs.

[db]

Thanks you very much!