Mothur and QIIME (trimming)

Hi All,

I used Mothur to trim a set of 454 sequences (used min and max length, qaverage=25, maxambig=0). I also used the Chimera Slayer and filter.seqs (on Mothur). I then imported a subset of these sequences into QIIME (I wanted to build some bootstrapped trees). I’ve noticed that the default trimming in QIIME includes removing any homopolymers >6 bp in length. I’m just wondering if anyone’s found that necessary, since I already cleaned up my files using the qual values on Mothur (and this additional step in Qiime is eliminating another ~3000 sequences from my already trimmed data set).

If anyone’s got insight into whether eliminating homopolymers is necessary, I’d appreciate it. I’ve looked at a couple articles, but haven’t really found what I’m looking for.

Thanks!

I used Mothur to trim a set of 454 sequences (used min and max length, qaverage=25, maxambig=0)

Just to be clear, this will do very little to actually remove sequencing errors as the reads come off the machine with an average quality score of 20. If you need to use quality scores, we have published data indicating that the best choice is qwindowaverage=35, qwindowsize=50. But really the best thing to use is trim.flows and shhh.flows. How to do this in mothur is available at http://www.mothur.org/wiki/454_SOP. Regardless of the method, we use a maxhomop value of 8. I would think 6 would remove some good stuff.

Pat