"From sequences to multi-omics: the computational side of biology"

Abstract :

Computational biology is a very broad discipline, seeking to build models for diverse types of experimental data and biological systems using methods from a wide range of mathematical and computational fields. Perhaps the most important task that computational biologists carry out is to frame biomedical problems as computational problems. This often means looking at a biological system in a new way, challenging current assumptions or theories about the relationships between parts of the system, or integrating different sources of information to make a more comprehensive model that better describes the complexity of the phenotypes.
In this seminar, I will provide an overview of how we modelled different biological systems using different types of input data to answer a wide variety of scientific questions. First, I will show how we can extract biological meaning from protein sequences leveraging bioinformatics and machine learning approaches to improve effector proteins characterization and prediction.
Then I will discuss how deep learning can model omics data to improve complex disease diagnosis.
Finally, I will show some preliminary results concerning the implementation of a novel approach at the crossroads between deep-learning and game theory to analyse multi-omics and multi-modal data to improve the identification of complex patterns of gene expression in plants during pathogen infection.

The seminar can also be followed via zoom:

https://inrae-fr.zoom.us/j/5785660130?omn=95873017305