Determining the different conformational states of a protein and the transition paths between them is key to fully understanding the relationship between biomolecular structure and function. I will discuss how ML models can learn a continuous conformational space representation from example structures produced by molecular dynamics simulations. I will then show how such representation, obtained via our software molearn, can be leveraged to predict putative protein transition states, or to generate conformations useful in the context of flexible protein-protein docking.
Matteo Degiacomi, obtained an MSc in Computer Science (2008) and a PhD in computational biophysics (2012) in Ecole Polytechnique Fédérale de Lausanne (EPFL). During his PhD supervised by Prof Matteo Dal Peraro he combined molecular dynamics simulations and particle swarm optimization to predict the assembly into complexes of the pore-forming toxin Aerolysin and the type-III secretion system’s basal body. In 2013 he joined the research groups of Prof Justin Benesch and Prof Dame Carol Robinson FRS in the University of Oxford. His research, funded by a Swiss National Science Foundation Early Postdoc Mobility Fellowship, focused on the development of new computational methods for the prediction of protein molecular assembly guided by ion mobility, cross-linking, SAXS and electron microscopy data, as well as their application to the study of small Heat Shock Proteins and protein-lipid interactions. In 2017 he obtained an EPSRC Junior Research Fellowship, allowing him to establish his independent research in Durham University, and in 2020 he was promoted to Associate Professor. In 2024 he moved to the University of Edinburgh, taking joint Reader position between the School of Informatics and the School of Chemistry.