Multi-scale modeling of microbial metabolic interactions

The National Institute of General Medical Sciences awarded the University of Vermont $406,167 on May 15, 2026, through the Biomedical Research and Research Training program (CFDA 93.859) to develop multi-scale computational models linking genes to metabolic interactions and in vivo microbial growth dynamics within the human microbiome, with focus on the infant gut microbiome as a model system.

Over five years through March 31, 2031, the recipient will integrate metagenomic sequencing data, measurements of microbial growth in defined communities, and quantification of microbial metabolic phenotypes across three analytical scales—the microbiome scale, community scale, and individual microbe scale—to enable predictive understanding of how competition and cross-feeding drive ecological dynamics in microbial communities. Work is performed in Burlington, Vermont.

Abstract

The human microbiome refers to the complex and dynamic communities of microbial organisms whose structure and function has been intricately linked to health maintenance and disease physiology. Metabolic interactions, such as competition and cross-feeding, drive ecological dynamics within these communities. Quantification of the microbial organisms and genes across the human microbiome has accelerated through the application of metagenomic sequencing. However, we still lack detailed mechanistic understanding linking genes to metabolic interactions and in vivo microbial growth dynamics. The complexity of these microbial ecosystems, consisting of thousands of microbial species and millions of distinct genes, necessitates computational approaches to organize data and knowledge. My lab's goal is to enable predictive understanding of microbiome dynamics through the development of computational models describing the multi-scale processes linking genes to metabolic functions and microbial interactions. Within this framework, we are focused on the infant GUT microbiome as an important model system. In early life, the GUT microbiome develops rapidly, has both acute and lasting impacts on human health, and has reduced complexity relative to the adult GUT microbiome. Over the next five years, my lab will pursue multi-scale understanding of microbial interactions in the infant GUT microbiome by integrating data across three different themes/scales. At the microbiome scale we will integrate metagenomic sequencing data into predictive models of in vivo microbial growth dynamics. At the community scale we will integrate measurements of microbial growth in simple defined communities to develop predictive models of microbial interactions. At the microbe scale we will integrate measurements of microbial growth and consumption/production of metabolites to predict individual microbial metabolic phenotypes from genotypes. Along with collecting and organizing data at each of these scales, we will focus on addressing three cross-cutting questions that bridge knowledge across scales. 1) can we identify the key microbial interactions driving in vivo growth dynamics? 2) which metabolic processes underlie key microbial interactions? 3) how do we effectively annotate the genes coding for these underlying metabolic processes? This research will provide mechanistic understanding of the ecology of the infant GUT microbiome as well as tools and approaches that are generally applicable to microbiome research. We will prioritize documenting and sharing our data and methods using fair data principles and open-source code to broadly enable the research community.