Using CCMetagen with custom database

[Former user]

I am trying to use a custom database consisting of all bacterial full length 16S sequences from SILVA 138 clustered at 97% assigned with de novo taxonomy when needed. I have the sample_out_kma.res from kma with the template looking like this: #Template ASV43690.1459 (.1459 corresponding to the length of the sequence). When running CCMetagen.py i get the following:

Traceback (most recent call last): File "/shared-nfs/TBNJ/BIOWIDE_2.0/scripts/envs/bin/CCMetagen.py", line 274, in <module> df = fParseKMA.populate_w_tax(df, ref_database, st, gt, ft, ot, ct, pt) File "/shared-nfs/TBNJ/BIOWIDE_2.0/scripts/envs/lib/python3.6/site-packages/ccmetagen/fParseKMA.py", line 90, in populate_w_tax match_info.Lineage = split_match[2] IndexError: list index out of range

Can i fix this by adding the taxonomy to the the sequence headers, as they only contains the taxid?

Best regards, Thomas

[ptlcc]

Hi Thomas

The taxids have to follow the same format as for the NCBI taxonomy browser.
The format of the header should look like this:

43690|ASV43690.1459 taxonomy description

For CCMetagen to recognise it, where “43690“ is the taxid.
You can alter this in *.name file, without having to reindex the entire thing.

Best,
Philip

[Thomas Bygh Nymann Jensen]

Hi Philip,

Thanks for the quick response.

Will do.

-Thomas

‌