R-CCS Cafe 第240回 第3部

For practical drug development it is necessary to efficiently simulate various and large-scale structural changes of target proteins. In all-atom molecular dynamics (AA-MD) simulations without any bias, it is difficult to reproduce the long-time (ms-s) dynamics of large-scale biomolecules. On the other hand, Coarse grained molecular dynamics (CG-MD) simulation, in which one degree of freedom for each residue is considered, has succeeded in reproducing various biological processes(dynamics) for many macro molecules qualitatively with lower calculation cost. However, CG-MD simulations remains several problems related to i) ambiguity of CG-parameters, ii) sampling efficiency, and iii) accuracy of force field. In this presentation, I would like to introduce an overview of my research and current efforts to overcome the above problems thorough development of new Coarse-Grained molecular dynamics simulations by using machine learning methods.