16S taxonomic analysis

[thermokarst]

@AndreaC

a microbial class goes from 20% to 60%!

To me that sounds like it has less to do with the database used, and more to do with the upstream processing steps appliedto your actual dataset, like QA/QC etc.

@AndreaC

Are such large differences justified by a different database?

Please see this resource, courtesy of @jwdebelius:

https://www.nature.com/articles/nbt.4229

@AndreaC

following your tutorials I don't seem to have made mistakes, are there any parameters or checks that I missed that could have compromised the results?

Please see @SoilRotifer's response to your question from a few weeks ago:

SILVA 16S database

I am referring to this drop-menu in the docs: [qiime-version-docs] Note this is specifically in reference to training nàive bayes classifiers, not necessarily Kraken 2 or other tools. In fact, I'd highly recommend reading the following papers about taxonomy assignments, and how to potentially improve them: Ziemski et al. 2021 Kaehler et al. 2019 Bokulich et al. 2018 Short answer, who knows. It really depends on what questions you are trying to answer, and what makes sen…