Marco De Vivo

Istituto Italiano di Tecnologia

Decoding Biochemical Complexity with Simulations and AI-Enhanced Sampling

Marco De Vivo*

Italian Institute of Technology

* Presenting author

I will present our efforts to use molecular simulations and AI-guided enhanced sampling to understand complex chemical processes that often involve functional conformational changes, which are difficult to capture and interpret unambiguously experimentally. Examples will include group II introns, self-splicing ribozymes that are evolutionarily and chemically related to the eukaryotic spliceosome. I will illustrate how we use simulations to delve into the mechanism by which the RNA core of splicing machineries, including the megadalton human spliceosome, operates at the atomic scale. In this regard, I will also discuss our current efforts to guide the design of RNA-targeted small molecules using free-energy perturbation calculations. I will also present our computational investigations of the mechanistic steps in nucleic acid polymerization catalyzed by DNA polymerases, which pose increasingly complex mechanistic questions and for which identifying the appropriate collective variable is inherently challenging. This exemplifies well the benefit of deep-learning-guided enhanced sampling of the configurational space. Overall, these examples demonstrate how atomistic molecular simulations can nowadays enable the prospective exploration of ultra-large, realistic multicomponent protein/nucleic acid systems.

All Sessions by Marco De Vivo

12:00 - 12:30
FUNCHAL 500 anos

Invited Lecture 28 - Marco De Vivo

Decoding Biochemical Complexity with Simulations and AI-Enhanced Sampling

Scroll to Top