ERROR: The MetaPhlAn2 taxonomic profile provided was not generated with the database version v30 . Please update your version of MetaPhlAn2 to v3.0

My yaml file installs humann=3.0.1.
And now I have: humann v3.0.1 and MetaPhlAn version 3.1.0 (25 Jul 2022).
But I still get the ERROR: The MetaPhlAn2 taxonomic profile provided was not generated with the database version v30 . Please update your version of MetaPhlAn2 to v3.0.error,
which doesn’t make much sense, since I have more updated version than V3.0

Is it because the humann has a check if the header has #mpa_v30_CHOCOPhlAn_201901 in it, and because my header is now #mpa_v31_CHOCOPhlAn_201901 I got an error?

Would love to see the tool gets updated for this.

Thank you.

Angel

oh I’ve figured out why. I should put humann=3.1.1 in my yaml file…

Glad you figured it out!

@franzosa can you elaborate more on a proper fix for this? We have this installed on our local HPC and are running into the same issue, and so I’ll need to point out the fix for our IT team.

Or, do you recommend that we update to the latest humann/metaphlan versions and do a clean database installation?

I sorted it out; the primary issue for us was a problem with the original 3.0.1 installation. The solution was updating to 3.1.1 and keeping a specific version of metaphlan3 tied to humann3 (currently using env modules for this), then explicitly downloading the databases in both cases to a shared spot on on our network file system and using the --nucleotide-database and --protein-database settings. Might switch to containers for future runs.

Now the main issue is speeding up DIAMOND with UniRef50 runs, but that is for another post…

Heya @cjfields :slight_smile:

I’m having a similar problem (i’m using conda to install humann), can you tell me what version of metaphlan3 you pinned to humann 3.1.1 that worked for you?

@AstrobioMike we used Metaphlan 3.1.0. Primary issue we ran into was pointing to a local download of the relevant databases (esp. the UniRef50/90).

1 Like