Project Details
Projekt
Systematic identification of phage-host cross-talk in the human gut microbiome
Subject Area
Microbial Ecology and Applied Microbiology
Bioinformatics and Theoretical Biology
Bioinformatics and Theoretical Biology
Term
since 2021
Project identifier
Deutsche Forschungsgemeinschaft (DFG) - Project number 464797012
Gut dysbacterioses, or imbalance in gut bacteria, are increasingly found to be associated with − and possibly cause for − a wide range of diseases, including metabolic syndrome, type II diabetes, fatty liver disease, and autoimmune diseases such as inflammatory bowel disease (IBD), or type I diabetes. Phages regulate bacterial abundance, diversity, and metabolisms in numerous ecosystems. While their role in the gut remains largely unexplored, it has been suggested that they may play a similar role there. Tens of thousands of phage and bacterial draft-quality metagenome assembled genomes (MAGs) have been assembled from human gut microbiomes. In order to understand the factors that shape the gut microbiome, we need better methods for linking these uncultured phages to their uncultured hosts, which to date are unknown for almost all phages discovered in the human gut. The set of known phage-host pairs (the "labeled data") is small and strongly biased towards a few cultured genera. This has largely prevented the application of machine learning methods. Our chief aim is to develop methods for the reliable identification of a large part of phage-host relationships in human gut microbiomes, including non-lysogenic phages. Objective I is to develop a statistical method that integrates the scores from five orthogonal sources of information while not requiring any labeled data at all. Objective II is to develop a statistical graphical model framework to reliably predict phage-host relationships from population-level high-throughput metagenomic co-abundance data and metatranscriptomic co-expression data. Objective III is to develop a viral tagging-based method to screen metagenomic libraries of phage particles against metagenomic bacterial libraries using single-cell RNA-seq to detect which phages infect which hosts. The methods are designed to be also sensitive to virulent, lytic phages, which current methods are mostly blind to. The elucidation of the phage-host interactions will be an important basis for the development of phage-based tools to precisely modulate the human microbiota by targeting bacterial taxa associated with dysbacteriosis and restore the healthy microbiome. These methods will be directly applicable to other human-associated and environmental microbiomes for which reference databases of MAGs of viruses and prokaryotes are available.
DFG Programme
Priority Programmes
