Error Running HUMAnN 3.0

Microbial community profiling HUMAnN
1

Hi all,

I am new to running metagenomic sequencing analysis. I have a .biom file that I have filtered to only include my baseline samples. I would like to look at the relative abundance of gene families and pathways using HUMAnN 3. Specifically I want to determine the difference in short-chain fatty acid producing pathways between my two groups of baseline samples.

I installed HUMAnN3.0 following this tutorial (GitHub - biobakery/humann: HUMAnN 3.0 is the next generation of HUMAnN 1.0 (HMP Unified Metabolic Analysis Network).). I am running the program using Ubuntu 22.04.02 LTS on a window PC. I believe I downloaded all of the necessary databases.

When I try to run HUMAnN3.0 using my .biom file:

humann --input acetate_BL_only_metagenome.biom --output acetate_humann

This gives me an “Index Error” issue (see attached below):

image
image1593×367 122 KB

My first guess is that I think that this error is likely due to something in my .biom file. So I ran biom validate-table with the same .biom file and I see that it is an unknown table type (see image below):

image
image1060×67 29.4 KB

When I ran ‘biom summarize-table’ it doesn’t look like I am missing any values (see image below):

image
image1514×2464 134 KB

In summary, I am not sure why my .biom file is not working as a valid input for HUMAnN3.0. Is there something I am missing or should check within my file to solve this index error? I am missing a step with the .biom file before it can be used as an input for HUMAnN3.0 - maybe I needed to run it through MetaPhlan?

Any and all advice would be greatly appreciated.

Best,
Nathan

2

Hi Nathan, Thank you for your detailed post. Sorry you are seeing such a confusing error message. It looks like HUMAnN has identified your file as a sam file which is not the correct format. Try running with the option --input-format biom to indicate your input file is of biom format.

If that does not work then double check your biom file only contains gene families for a single sample. If it does, then I wonder if your biom file is of a new format? If so, let me know the format so we can work on supporting it in a future release.

Thank you,
Lauren