Convert mothur classify.seqs output to BIOM format

[Petr Kozyrev]

I got a lot of problems trying to convert output of mothur classify.seqs to BIOM format so that I can import data into phyloseq R package. I have FASTA file with paired-end reads already being merged with PEAR program.

I aligned my sequences to SILVA database with this command

mothur > align.seqs(fasta=sample.fasta, reference=silva.bacteria.fasta, processors=4, flip=t)

and then I classified it using the same database

mothur > classify.seqs(fasta=sample.align, reference=silva.bacteria.fasta, taxonomy=silva.bacteria.silva.tax, cutoff=80)

After this procedure I got different files: "sample.summary", "sample.silva.wang.taxonomy" and "sample.silva.wang.tax.summary". But I don't know how to import them into phyloseq R package.

I've read here https://github.com/joey711/phyloseq/issues/245 about shared file from which I can create BIOM file, but I don't have .list or .group files for make.shared command.

Can someone help?

Sincerely yours,

Petr

[thermophile]

You have only classified your sequences, you haven't clustered them into OTUs which generates your list and shared files. Have you tried following the mothur how to?

https://www.mothur.org/wiki/MiSeq_SOP

[Petr Kozyrev]

Thank you for your quick response!

I didn't mention that I used RDPTools classifier for my purposes earlier, but then some samples were classified badly. So I decided to check, if RDPTools classifier is not very good, and try mothur classifier. To achieve my goal I wanted to omit some steps in mothur SOP so that I can compare only classifiers in both pipelines. But now I realize it wasn't a really good idea.

I'll try to follow this SOP step by step now.

Petr