I was aware there is another similar request in a different group but just wanted to bring this to your attention so I repeat it here:
would it be possible to implement the uclust otu picking and taxonomy assignment using uclust and/or usearch61 or usearch (latest version). It turns out to be better than the RDP classifier for my data.
The short answer is no, we do not plan to add these methods.
Uclust is only licensed for free use with qiime1. Usearch is not free either, unless if you use the 32-bit version.
For sequence clustering we have wrapped vsearch, which is a free tool that replicates the functionality of usearch. See the
q2-vsearch plugin for more details on the OTU clustering methods available.
For taxonomy classification, we have also implemented a vsearch-based LCA taxonomy classifier in
q2-feature-classifier. The uclust-based classifier (which we implemented in qiime1 using the same alignment + LCA consensus taxonomy assignment approach) is no more accurate than vsearch classifier and other methods; in fact, it is less accurate than the naive bayes classifier in
q2-feature-classifier according to our benchmarks.
Thanks for the suggestion!
This topic was automatically closed 31 days after the last reply. New replies are no longer allowed.