Video talks on 16S data analysis posted.

URMAP ultra-fast read mapper posted (paper).

~20% of taxonomy annotations in SILVA and Greengenes are wrong (paper).

Taxonomy prediction is <50% accurate for 16S V4 sequences (paper).

97% OTU threshold is wrong for species, should be 99% for full-length 16S, 100% V4 (paper).

Westcott & Schloss's definition of OTUs

In Westcott and Schloss 2017, the authors propose the following requirements for OTUs constructed from 16S sequences.

1. All pairs of sequences in the same OTU must be at least 97% similar.

2. All pairs of sequences in different OTUs must be less than 97% similar.

However, with real data is it usually impossible to satisfy these requirements, as illustated by a simple example.

The authors do not discuss this problem in the paper. They implicitly acknowledge the issue by proposing that the clusters with the highest value of the MCC score are the best OTUs. I disagree -- see discussion of the example.