Recorded talk: Video

Speaker: Sergei Grudinin, CNRS, France.

Title: Structural bioinformatics in the era of AlphaFold

Registration: https://lu-se.zoom.us/meeting/register/u50ud-qvrTItEtO77Pw88sn-UOi5G-c2tkjJ

Abstract:

While the problem of determining how a protein folds in three dimensions (3D) is essentially solved, accessing protein motions and interactions at physiological conditions is becoming more central than ever before. Indeed, proteins are flexible biological objects, constantly moving and changing their shape to interact with their environment and cellular partners. This inherent flexibility is highly relevant for protein functioning. Experimentally, it is very challenging to observe proteins directly in action at high resolution, and we mostly have access to isolated clusters of “snapshots" (conformations) representative of a few functional states. I will give an overview of the recent progress in protein structure prediction. I will then present physics-based and machine-learning models developed by our team to predict protein flexibility and their functional motions. These models have proven to be very useful in predicting protein dynamics and interactions up to the cellular level or extrapolating proteins' functional motions. I will demonstrate applications allowing the construction of multi-level representations of protein flexibility and integrative algorithms driven by low-resolution experimental observations, such as small-angle scattering.

Bio:

 Sergei Grudinin graduated from the Moscow Institute of Physics and Technology (MIPT) in 2002. He did his Master's project on MD simulations of membrane proteins in Forschungzentrum Juelich, where he continued it as a Ph.D. thesis. In 2006, after the Ph.D. defense, he extended the activities to method development, and later moved to Inria Grenoble, France, in 2007 and became a permanent CNRS scientist in 2009. During the last few years, his main research interests were the development of physics-based and data-driven algorithms for structural bioinformatics. They include novel methods for protein-protein and protein-ligand docking, including those accelerated with the Fast Fourier Transform; algorithms to study and predict symmetrical systems; algorithms to describe molecular flexibility; methods for integrative structural biology, e.g., those related to SAXS, SANS, and X-links measurements; applications of convex optimization to protein structure prediction; development of deep convolutional neural networks for structural bioinformatics; and graph- and tessellation-based geometric learning approaches.