Are humann_regroup_table (utility_mapping) command and flag –utility-database in HUMANnN v4.0.0.alpha.1 doing the same thing?

Microbial community profiling HUMAnN
1

Hi all,

I have run HUMAnN (v4.0.0.alpha.1) on a dataset using the flag –utility-database:

–utility-database <utility_mapping_database>
directory containing the utility database
[DEFAULT: /usr/local/lib/python3.10/site-packages/humann/data/utility_DEMO]

Providing the folder in which the utility_database was downloaded.

I was wondering if using these settings coincide with the usage of humann_regroup_table, since from what I read in the github page relies on the same datadabse. This doubt came from the fact that running humann_regroup_table on my actual results just gives a table containing UNGROUPED and UNINTEGRATED entries.

Did I interpretate this correctly? Am I doing something wrong?

Thanks a lot

2

The --utility-database covers all the database files HUMAnN uses that aren’t big sequence files (e.g. mappings from gene IDs to reaction IDs). Normally you only refer to --utility-database when doing an initial download OR if you moved the download and need to update your config. Otherwise scripts like regroup_table look to the config automatically to see where the utility database lives and load files accordingly.

Right now it looks like the only utility database you have is the demo one, so you might just need to execute a download of the full utility mapping database?

1 Like
3

Indeed, I think I solved the proble updating the database and relative path. Thank you!