ASVs or OTUs in Taxanomy Classification/Analysis?

[Mehrdad (Meha)]

Hello active admins,

I have got a question concerning taxonomy. Up to DADA2, we were talking about ASVs, which are reliable rather than OTUs. My shining question is raised on this case that why OTU is put forwarded over and over again in SILVA and GreenGen database when taxonomy analyzing step?

https://docs.qiime2.org/2019.1/data-resources/

I used DADA2 plugin, absolutely I created ASVs, not OTUs! How the outputs from by DADA2 tool are compatible with downstream processes, such as taxonomy analysis? It sounds that is a heterogeneous condition. Does not it make sense?

How could you justify it?

Thanks

[Nicholas_Bokulich]

Hi @Mehrdad,
There is no incompatibility here, you are overthinking things.

The SILVA and Greengenes databases mention OTUs because those reference sequences are clustered (at 99% similarity), basically to remove replicate/similar sequences from the database. This is done to make those databases more space and memory efficient. This has no bearing on how those databases can be used.

So you can use those databases whether your data consist of ASVs or OTUs or raw sequences.