Abstract:
The ability to rapidly learn from high-dimensional data to make reliable predictions about the future is crucial to life. This could be a fly avoiding predators, or the retina processing terabytes of data guiding complex human actions. Modern day artificial intelligence (AI) aims to mimic this fidelity and has been successful in many domains. It is tempting to ask if AI could also be used to understand and predict the dynamics of complex molecules with millions of atoms. In this talk I will show that certain flavors of AI can indeed help us understand generic molecular dynamics and also predict it even in situations with arbitrary long memories. However this requires close integration of AI with old and new ideas in statistical mechanics. I will talk about some such methods developed by my group (1-3). I will demonstrate the methods on different problems, where we predict mechanisms at timescales much longer than milliseconds while keeping all-atom/femtosecond resolution. These include ligand dissociation from flexible protein/RNA and crystal nucleation with competing polymorphs. I will conclude by discussing some generic challenges and solutions regarding reliability, interpretability and extrapolative powers of AI when used in molecular simulations.
References:
1. Wang, Y., Ribeiro, J.M.L. & Tiwary, P. Past–future information bottleneck for sampling molecular reaction coordinate simultaneously with thermodynamics and kinetics. Nat. Commun. 10, 3573 (2019). https://doi.org/10.1038/s41467-019-11405-4
2. Tsai, S.T, Kuo, E.J. & Tiwary, P. Learning Molecular Dynamics with Simple Language Model built upon Long Short-Term Memory Neural Network. Nat. Commun. 11, 5115 (2020). https://doi.org/10.1038/s41467-020-18959-8
3. Wang, Y., Ribeiro, J.M.L. and Tiwary, P. Machine learning approaches for analyzing and enhancing molecular dynamics simulations. Curr. Op. Sruc. Bio., 61, 139 (2020). https://doi.org/10.1016/j.sbi.2019.12.016