I am currently working on a gut microbiome project focused on the gut microbiome changes in a mouse EAE model. We have data processed with QIIME and most of our analysis has been on R up to this point. I'm familiar with the use of metagenomic screens and sequencing to infer the metabolomics of a microbiota, but we only have 16S abundance data. I wanted to know if there were any cited/accepted methods for pairing bacterial abundance at the species or above level with sequenced genomes/pathways to infer relative levels of "metabolic activity". I am aware that KEGG data can be leveraged for similar purposes: Article Metabolic-network-driven analysis of bacterial ecological strategies
We are currently focused on network analysis of abundance data but are interested in the metabolite aspect as well.
You should take a look at PiCrust (https://www.nature.com/articles/nbt.2676)
Yes, I think PICRUSt is indeed the most used method up to date. However in order to work with it, your sequence picking must be based on the Green Genes Database, which is not the largest available now and not updated for years now...
I would be really interested if there is any tool that works with a Silva database matching.
Tax4Fun does the same but uses Silva as reference.
http://tax4fun.gobics.de/
Oh, by the way, there is already a PICRUSt 2.0 beta release available (https://github.com/picrust/picrust2). The new version has overcome some previous limitations: it is database agnostic, allows biological sequences and the predictions are based on the latest IMG genomes.
- Badges
- Science topic
- Similar topics
- Biological Science
- Microbiology