Biophysical Society Thematic Meeting | Hamburg 2022

Biophysics at the Dawn of Exascale Computers

Tuesday Speaker Abstracts

DEVELOPMENT OF PREDICTIVE APPROACHES FOR BIOMOLECULAR ASSOCIATION KINETICS Karen Palacio-Rodriguez 1,2 ; Hadrien Vroylandt 3 ; Lukas S Stelzl 4 ; Gerhard Hummer 5,6 ; Pilar Cossio 2,7 ; Fabio Pietrucci 1 ; 1 Sorbonne Université, Institut de Minéralogie, de Physique des Matériaux et de Cosmochimie, IMPMC, Paris, France 2 University of Antioquia, Biophysics of Tropical Diseases Max Planck Tandem Group, Medellín, Colombia 3 Sorbonne Université, Institut des Sciences du Calcul et des Données, ISCD, Paris, France 4 Johannes Gutenberg University Mainz, Faculty of Biology, Mainz, Germany 5 Max Planck Institute of Biophysics, Department of Theoretical Biophysics, Frankfurt, Germany 6 Goethe University Frankfurt, Institute for Biophysics, Frankfurt, Germany 7 Flatiron Institute, Center for Computational Mathematics, New York, NY, USA Atomistic computer simulations of rare events have three paramount goals: predicting detailed mechanisms, free energy landscapes, and kinetic rates. In real-life applications, all of these tasks are cumbersome and require intensive human and computer effort, especially the calculation of rates. We developed two efficient methodologies for the prediction of transition rates from molecular dynamics simulations. Both strategies only require sets of short simulations, which allows exploiting the parallel capabilities of current supercomputers. On one side, transition path sampling trajectories are the golden standard to access mechanistic information: we demonstrate that they also encode accurate thermodynamic and kinetic information, that can be extracted by training a data-driven Langevin model of the dynamics projected on a collective variable [1]. We use fullerene dimers as a proxy system to protein-protein interactions and recover free energies, position-dependent diffusion coefficients, and rates. On the other side, we use metadynamics, an enhanced sampling technique that allows accelerating the sampling of rare events but distorts the dynamics. We overcome this limitation by developing a method based on Kramers’ theory for calculating the barrier-crossing rate when a time-dependent bias is added to the system [2]. We tested this method in a double-well potential and in the fullerene dimers, showing that we are able to extract the rate and measure at the same time the quality of the collective variables. Finally, we apply the method to a complex protein-ligand interaction (CDK2-03K) reproducing the experimental unbinding rate up to an order of magnitude discrepancy. Overall, these new theoretical tools make efficient use of computing resources providing simple procedures to accurately predict kinetic rates and could be suitable for applications far beyond the field of biomolecular association.References:1. Palacio-Rodriguez, K., & Pietrucci, F. (2021). Free energy landscapes, diffusion coefficients and kinetic rates from transition paths. arXiv preprint arXiv:2106.05415.2. Palacio-Rodriguez, K., Vroylandt, H., Stelzl, L. S., Pietrucci, F., Hummer, G., & Cossio, P. (2021). Transition rates, survival probabilities, and quality of bias from time- dependent biased simulations. arXiv preprint arXiv:2109.11360.

12