I’m new to QIIME 2 and still learning about functional prediction. I have a theoretical question about how to interpret MetaCyc and KEGG pathway outputs from PICRUSt2.
From what I understand:
MetaCyc focuses mainly on microbial metabolic pathways.
KEGG includes both microbial and host-related (eukaryotic) pathways.
If I’m analyzing 16S rRNA-based fecal microbiome data (so there are no human genes in the input), should the KEGG pathways still be considered as containing both host and microbial pathways?
In my KEGG output, I see pathways such as Alcoholic liver disease, Huntington’s disease, and Carcinoma. Does this mean there might be an error in my analysis, or can microbial genes still be annotated to these host-related pathways because of shared enzymes or homologous gene functions?
I’d really appreciate clarification on how to correctly interpret these KEGG results for 16S-based analyses.
I'm not an expert on Picrust2, but I think this discussion may be useful:
I actually wouldn't recommend using PICRUSt2 for KEGG pathways unless you have a subscription to KEGG and can replace the pathway file with one of your own. We don't have these pathways output as default for this reason, and you can see a little more information on this here.
To answer your question more directly:
should the KEGG pathways still be considered as containing both host and microbial pathways?
No. The pathways are predicted by Picrust2 using amplicon data that is all microbial (presumably), so the pathways are all microbial (presumably).
I’d really appreciate clarification on how to correctly interpret these KEGG results for 16S-based analyses.
If this was my project, I'm not sure I would use KEGG at all...
