- Meeting ID: 867 6409 6440
- passcode: 149120
13h00 - 13h50 – Clémence Frioux (INRIA)
Machine reasoning, dimensionality reduction, and numerical modelling for exploring microbial community metabolism
ABSTRACT Microbial communities and microbiomes are complex ecosystems inhabited by up to hundreds or thousands of microbial populations that interact closely with each other, their environment, and, in some cases, a host. Because microbiomes provide essential services to their ecosystems, gaining a deeper understanding of them has broad applications in health, agroecology, and industry. Computational biology plays a key role in processing, analysing, and modelling the vast experimental data generated on microbial communities, of ten through culture-independent approaches based on DNA sequencing. I will present several contributions in this field, ranging from the analysis and dimensionality reduction of microbiome composition data to mechanistic models of microbial community dynamics. The concept of metabolic network is central to my work, as it enables the inference of microbial phenotypes by linking genetic sequences to metabolic reactions that transform molecules. By leveraging this concept and computational formalisms based on numerical methods or machine reasoning, we can generate predictions about microbiome functions, the interactions within them, and, more broadly, transform experimental data into simplified mechanistic models of the underlying biological systems.
Dernière modification le 05/12/2025