Biophysics in latent space

Abstract:

Many phenomena in biology are considered too complicated or too contingent to be captured by predictive theories similar to what is done in physics. But complex systems theory has taught us that simple, higher level laws with few effective parameters can emerge from the interaction of small scale components. As biology is becoming more and more quantitative, one can use a combination of first-principle theoretical modelling with simple machine learning techniques to build accurate and tractable theories of biological dynamics. Those dynamics can often be best understood in (abstract) latent spaces, giving « physics-like » intuition.

I will illustrate the power of such approaches on a couple of biological examples, with a special focus on the dynamics of the adaptive immune system (T cells response). Our approach leads to applications in cancer immunotherapy that I will briefly describe.

Bio:

Paul François is a Professor of Bioinformatics at the Université de Montréal and an associate member of MILA, an AI research institute in Quebec. He earned his PhD from Université Paris in 2005, and today is a leading expert in theoretical and computational biophysics. His team applies machine learning approaches to explore an array of biophysics topics, including systems biology, developmental biology, evolution, and quantitative immunology. Among other accolades, Prof. François was awarded the Rutherford Memorial Medal in Physics by the Royal Society of Canada in 2019 and the CAP Herzberg Medal in 2017.

Learn More:

See Paul's faculty research page here

See the François group website here