Friday, June 7, 2019 3:00 pm – 4:30 pm
SalleSC 10.01 Université de Montpellier –Campus Triolet
Machine learning for the discovery of new physiology from omics data
Genome-scale molecular profiles are becoming accessible for many biomedical and clinical research projects. The discovery of physiological processes from this data can be attempted using models that are based onsingle measurements, or linear combinations thereof.
The underlying biology reality is often likely to be more complex than this, but limited available data and limited computational resources make it difficult to go beyond these simple models while preserving statistical power and interpretability of the results.In my talk, I will discuss two cases of new and carefully calibrated computational approaches that allowed us to discover and validate previously unknown associations between transcriptomic data and biomedically relevant phenotypes. Both models are highly interpretable and generic enough to be applied to a wide range of transcriptome analysis scenarios.
Benno Schwikowski is head of the Systems Biology Group at the Institut Pasteur, which develops computational tools to discover causal pathways in complex diseases.