Rename_table doesn't work with uniref90

Microbial community profiling HUMAnN
1

I face an error when trying to attach names to features
I used uniref90 to run humann.

humann2 --input kclsho.fq.gz --nucleotide-database /resources/chocophlan/ --protein-database resources/uniref90

Then…

humann2_rename_table --input kclsho_genefamilies.tsv --output kclsho_genefamilies-NAMES.tsv --names uniref90

And here is the error…

usage: humann2_rename_table [-h] [-i INPUT]
[-n {infogo1000,metacyc-rxn,kegg-module,ec,go,metacyc-pwy,pfam,eggnog,uniref50,kegg-pathway,kegg-orthology}]
[-c CUSTOM] [-s] [-o OUTPUT]
humann2_rename_table: error: argument -n/–names: invalid choice: ‘uniref90’ (choose from ‘infogo1000’, ‘metacyc-rxn’, ‘kegg-module’, ‘ec’, ‘go’, ‘metacyc-pwy’, ‘pfam’, ‘eggnog’, ‘uniref50’, ‘kegg-pathway’, ‘kegg-orthology’)

When I try using any of the suggested names it renames 0, as all features start from “'UniRef90_” which seems incompatible.

Any hints, please?

2

Hello, Sorry for the confusion. The uniref90 name mapping files are large so we can’t include them in the pypi/conda package. You can download them with the other large utility files with the following command:

$ humann_databases --download utility_mapping full $DIR (replace $DIR with the location to store the files)

Once these files are downloaded you will be able to select uniref90. Sorry again for any confusion.

Thank you,
Lauren

3

Thank you very much for the swift reply, Lauren.
I will give it a go.

Max