qseq files versus sequence.txt files

[drio]

Does anyone know what how GApipeline filters the reads that go into the final s_#_sequence.txt files ? I would like to know what criteria they use.
I couldn't find it in the documentation.

[kmcarr]

Here is page 85 from the Pipeline User's Guide which describes the filtering, including what the default method and threshold are.

[drio]

Thanks!

I am going to try a little experiment. Map the raw data (qseqs) compared to using
the filtered data (sequence).

Intuitively the default filter seems a little bit aggressive, specially for >36bp reads.
I don't see any problem with aligning the raw data as soon as the aligner takes into
account the qualities. Presumably reads that would had been removed will have low
qualities on the first 25bp.

What are your thoughts?

[lcollado]

I find this thread interesting as we recently got some 50bp data. Should we use the default parameters, which I guess are ideal for 36bp reads, or use some custom ones?