PICRUSt in rat-derived data?

[Karol Perlejewski]

Hi everyone, 

I would like to implement PICRUSt in my research but I have some doubts. 

In short.
We have 16S rRNA sequencing data generated from stool analysis collected from rats treated in three different ways. I read an article: "Predictive functional profiling of microbial communities using 16S rRNA marker gene sequences" where at is stated  that "PICRUSt recaptures key findings from the Human Microbiome Project and accurately predicts the abundance of gene families in host-associated and environmental communities, with quantifiable uncertainty."

Does it mean that you can use PICRUSt and interpret its results only in human-based models?

On the other hand you stated following: "We applied PICRUSt to a range of datasets from humans, soils, other mammalian guts and the hyper-diverse and under explored Guerrero Negro microbial mat".

So it is possible to apply PICRUSt for datasets from rats.
My question is how to do it technically? Should I speccify something during categorize_by_function.py step (KEGG pathway specific forrats)? If I use parameter " -m KEGG_Pathways" will it make calculations for KEGG human pathways?

I have found KEGG pathway maps - Rattus norvegicus (rat)
https://www.genome.jp/kegg-bin/show_organism?menu_type=pathway_maps&org=rno
 Should I somehow implement this?

[Gavin Douglas]

Hi there,

First off I should mention that PICRUSt was updated and validated with new approaches in this manuscript from 2020: https://www.nature.com/articles/s41587-020-0548-6?proof=t

Does it mean that you can use PICRUSt and interpret its results only in human-based models?

On the other hand you stated following: "We applied PICRUSt to a range of datasets from humans, soils, other mammalian guts and the hyper-diverse and under explored Guerrero Negro microbial mat".

You can apply PICRUSt to microbiome samples from rat stool samples and they should be more accurate than expected by chance. However, regardless of the environment, you should be aware that the predictions are not highly specific and only have resolution at the level of the 16S sequence: there is a lot of pangenome diversity that PICRUSt can’t capture.

Regarding KEGG pathways in PICRUSt1 - the pathways are actually all based on microbial gene families that can share similarity with genes in humans (or rats). So these functions are all based on the microbiome, not on the host, so you definitely don’t want to try to infer anything about pathways expressed by rats based on 16S data alone.

Cheers,

Gavin

--
You received this message because you are subscribed to the Google Groups "picrust-users" group.
To unsubscribe from this group and stop receiving emails from it, send an email to picrust-user...@googlegroups.com.
To view this discussion on the web visit https://groups.google.com/d/msgid/picrust-users/ad425145-d388-4969-a3fc-333f6590b5d7n%40googlegroups.com.