Hi
I have a simple question. Most of the papers I am reading use a “SINGLE” primer (say V6) and then the resulting downstream data is clustered into OTUs (I said most…)
Suppose, I have a mock community of 10 bacterial strains (assume ALL of them have ONE 16s operon) and I subsequently use 3 different primer sets (V3, V4-5 and V6), sequence it and cluster it into OTUs, would it be fair to assume that I would obtain 30 OTUs using a pairwise, average linkage algorithm?
i.e., Does multiplexing multiple primer sets for sequencing lead to OTU inflation downstream?
Thanks for the favour of a reply!
I think you mean a single primer set to sequence a single region. This is my preferred approach. I’ve been a part of projects that thought it would be grand to sequence 2 or 3 regions and in fact, this is part of the design behind the IonTorrent approach. It’s a headache because it’s often very difficult to synthesize the results across regions when the primers all have different biases. You would definitely need to process the three regions separately since there wouldn’t be any overlap between the regions. Stick with one region and sequence 3 times the number of samples or get 3 times the data from the region. And if you’re doing MiSeq, sequence the V4 region…
http://blog.mothur.org/2014/09/11/Why-such-a-large-distance-matrix%3F/
Pat