pipeline understanding from assign_taxonomy.py to biom add-metadata

[Guerrino Macori]

Hello Qiime users,

I would like to ask you a clarification about the command and the output of the following stage of the pipeline.

In order to assign the taxonomy to each sequence I used this command:

assign_taxonomy.py -i FILE.fasta -r 97_otus.fasta -t 97_otu_taxonomy.txt 

That generates the following files:

FILE_tax_assignments.log

FILE_tax_assignments.txt

Now, I would like to generate the final.biom but, using the following command:

biom add-metadata -i FILE_tax_assignments.txt -o final.biom -m FILE_Map.txt 

I'm receiving an error:

ValueError: could not convert string to float: Unassigned

I converted the file in biom as well but with the same kind of error. Can somebody explain me better what's the error in the command please?

Thanks in advance!

[Daniel McDonald]

Hi Guerrino,

The biom add-metadata command requires a few additional pieces of information. First, you need to specify your OTU table as "-i". Second, your taxonomy assignments need to be passed to "--observation-metadata-fp". Third, you need to indicate what columns within your taxonomy assignment file you'd like to load into the BIOM table, and this can be done with "--observation-header OTUID,taxonomy". Last, your taxonomy assignments are (likely) semicolon separated, so you need to specify "--sc-pipe-separated taxonomy". I believe that should resolve your issue.

$ biom add-metadata -i <your_otu_table.biom> -o final.biom --observation-metadata-fp FILE_tax_assignments.txt --observation-header OTUID,taxonomy --sc-pipe-separated taxonomy

Best,

Daniel

[Guerrino Macori]

Dear Daniel,

Thanks a lot for your explanation. I proceed with the following command:

$ biom add-metadata -i ~/merged.biom -o ~/final.biom --observation-metadata-fp ~/FILE_tax_assignments.txt --observation-header OTUID,taxonomy --sc-pipe-separated

This generated the file final.biom. I am trying to understand if the file generated is correct and if I can proceed with it for the creation of the graphs and downstream analysis.

Thanks in advance again

[Daniel McDonald]

Great!

Best,

Daniel