Hi,
I would like to filter OTUs by their relative abundance based on different criteria, e.g.
- OTUs occurring => 50 % of the groups (works with filter.shared -minpercent=50 I guess),
- occurring < 10 % of the groups,
- and everything in between those (10 % < OTUs < 50 %).
I haven’t found any option to get the last two criteria… Is there currently any? I thought it would be possible to save the discarded OTUs, in a similar .filter.shared file, thus being able to ‘work around’ and getting the above mentioned criteria fulfilled.
If not currently existing, maybe could be interesting implementing flags making this possible in a future release?
Thanks,
J
- OTUs occurring => 50 % of the groups (works with filter.shared -minpercent=50 I guess)
filter.shared(shared=yourSharedFile, minpercentsamples=50)
- occurring < 10 % of the groups,
filter.shared(shared=yourSharedFile, minpercentsamples=10) - get OTUs with at least 10%
list.otulabels(shared=current) - names of OTUs you don’t want
remove.otulabels(shared=originalSharedFile, accnos=listOtuLabelsOutput) - remove all OTUs with sequence present in more than 10% of the samples, leaving only those with < 10%
- and everything in between those (10 % < OTUs < 50 %).
filter.shared(shared=yourSharedFile, minpercentsamples=50) - get OTUs with at least 50%
list.otulabels(shared=current) - names of OTUs you don’t want
remove.otulabels(shared=originalSharedFile, accnos=listOtuLabelsOutput) - remove all OTUs with sequence present in more than 50% of the samples, leaving only those with < 50%
filter.shared(shared=current, minpercentsamples=10) - get OTUs with at least 10%
Neat, exactly how I imagine it to work even if I didn’t sort it out myself! Big thanks!
/J