2023-05-15 19:35:41
Paul François is a big fan of science fiction. But not just any. “What I like is science fiction turned towards biology and otherness, which depicts our interactions with extraterrestrial societies. If you can imagine something in the living world, it must exist somewhere,” he says.
Exploring the game of possibilities: this is the passion that animates the biophysicist even in his laboratory, where he is busy creating mathematical models using computer tools to better understand biology. In short, he “provides simple answers to complicated problems”.
The very course of the new full professor of the Department of Biochemistry and Molecular Medicine of the University of Montreal comes down entirely to this quest for simplicity in complex terrain.
“At first, I did math, but it wasn’t concrete enough. Then I turned to physics to find direct applications. My new functions at the Faculty of Medicine allow me to open up to medical biology and my affiliation to Mila [Institut québécois d’intelligence artificielle] puts me in interaction with machine learning. Recent progress in artificial intelligence can teach us regarding biological phenomena and lead us to better understand certain diseases”, explains the man who obtained an engineering degree at the École polytechnique de France, a doctorate in theoretical physics at the École Normale Supérieure (Paris) and a post-doctorate in bioinformatics at Rockefeller University (New York).
Several interdisciplinary and transdisciplinary studies conducted at UdeM in quantitative sciences – mathematics, physics, machine learning, biology, medicine – convinced the former professor of physics at McGill University to make the leap to the mountain campus, within new Biomedical Innovation Centerwhich brings together researchers from the five basic science departments of the UdeM Faculty of Medicine.
Deciphering the language of immune cells
A follower of “cross-fertilization” between disciplines, Paul François continues his research on adaptive immunity, which he began more than 10 years ago. How does the immune system recognize pathogens? What information does it encode to trigger an immune response? A scientific approach resulting from interactions between colleagues has brought out new mathematical models. “Initially, I did not understand the model of my collaborator Grégoire Altan-Bonnet, who heads the immunodynamics laboratory at the National Cancer Institute in the United States. [NCI]. By talking to him, I found a way to simplify it to make the data more consistent,” explains Paul François.
In collaboration with Grégoire Altan-Bonnet’s group, his team now uses a robotic platform combined with machine learning and mathematical models that quantify the immune response. “With clinicians, including Naomi Taylor from the NCI, we make mathematical predictions regarding how to adapt cancer immunotherapies. Our research might quickly lead to concrete applications,” he hopes.
Japanese fish and embryonic development
Another favorite subject of the researcher is the mathematical modeling of the Embryonic development. “You should know that a genetic oscillator [appelé horloge de segmentation] directs the formation of the segments, one vertebra at a time, without us really knowing how. I am working to describe this phenomenon from a mathematical point of view, in particular from a study on a Japanese fish which develops under different temperatures, he indicates. Our collaborators at the European Molecular Biology Laboratory vary the temperature cycles in order to disrupt the formation of his vertebrae. These manipulations allow us to understand how the body takes shape, what disturbs it and how it reacts to disturbances. If he does not resist, diseases appear.
This research is a prelude to other experiments on cell differentiation from embryos recreated in Petri dishes, for example. “If we manage to control embryonic development, we will be able to test treatments more easily,” underlines the scientist.
Machine learning at the service of biology
A subgroup of his laboratory is finally working on evolution algorithms. “We have designed computer tools to simulate the evolution of biological networks which we have applied, among other things, to the control of cell size,” he says. These models show us that if senescent cells are too big, it is because they are dividing too late or too early; the oscillator that controls the cell cycle is faulty.”
Professor François hopes that all this work will encourage his students to break down the barriers between disciplines. “I find that young people today are much more open to interdisciplinarity. My role is to help them develop their scientific tastes and get them to do concrete things,” he concludes.
1684199871
#Paul #François #science #fiction #fundamental #mechanisms #biology
