how to train taxonomy classifier when primers are unknown?

[ErikaGanda]

On that same note, what are the implications of me using a full-length 16S classifier with data of an unknown subregion of 16S - 151 bp?

Backstory: Someone gave me data and no primer sequences and I am not sure what the best course of action is. Was thinking about using the full-length pre-trained classifier and hoping it would match the region I have, would that work?

[Nicholas_Bokulich]

Welcome back @ErikaGanda!

@ErikaGanda

Was thinking about using the full-length pre-trained classifier and hoping it would match the region I have, would that work?

As long as your amplicon falls somewhere within the full-length 16S, using the full-length 16S seqs to train a classifier is a fine idea, and exactly what I would do.

Good luck!

[ErikaGanda]

Thank you @Nicholas_Bokulich! Good to be back doing some fun data analysis =)

[ben (benjamin w.)]

Can you query the barcode file or their core? It seems a little weird that the V region would be unknown given the relatively standardized primers that are used. However, some labs use their own, good luck. Ben