Using cross-species co-expression to predict metabolic interactions in microbiomes.

Area of Science:

• Microbial Ecology
• Systems Biology
• Bioinformatics

Background:

• Metabolic interactions govern microbial community structure and function.
• Predicting these interactions computationally is crucial but often lacks mechanistic insight.
• Existing tools often miss underlying metabolic pathways, hindering experimental validation.

Purpose of the Study:

• To develop and validate a novel approach using cross-species co-expression to predict microbial interactions.
• To identify specific metabolic pathways involved in competition, cross-feeding, and specialized interactions.
• To assess the potential for discovering novel gene functions through interaction prediction.

Main Methods:

• Applied cross-species co-expression analysis to microbial co-culture RNA-sequencing data.
• Utilized Mucin and Diet-based Minimal Microbiome (MDb-MM) and Hitchhikers of the Rhizosphere (THOR) datasets.
• Investigated gene and pathway co-expression patterns to infer interaction types.

Main Results:

• Successfully predicted pathways involved in resource competition within the MDb-MM dataset.
• Identified links between specialized functions, such as antibiotic and multidrug efflux systems in the THOR dataset.
• Provided evidence for siderophore co-expression driving interactions in a specific microbial consortium.

Conclusions:

• Cross-species co-expression is a feasible and data-driven method for predicting microbial interactions and underlying pathways.
• This approach offers a valuable alternative to complex model-building, reducing bias.
• The method facilitates the discovery of novel gene functions and informs microbiome engineering.