[en] Classical intermolecular potentials typically require an extensive parametrization procedure for any
new compound considered. To do away with prior parametrization, we propose a combination of
physics-based potentials with machine learning (ML), coined IPML, which is transferable across
small neutral organic and biologically relevant molecules. ML models provide on-the-fly predictions
for environment-dependent local atomic properties: electrostatic multipole coefficients (significant
error reduction compared to previously reported), the population and decay rate of valence atomic
densities, and polarizabilities across conformations and chemical compositions of H, C, N, and O
atoms. These parameters enable accurate calculations of intermolecular contributions—electrostatics,
charge penetration, repulsion, induction/polarization, and many-body dispersion. Unlike other potentials,
this model is transferable in its ability to handle new molecules and conformations without
explicit prior parametrization: All local atomic properties are predicted from ML, leaving only eight
global parameters—optimized once and for all across compounds.We validate IPML on various gasphase
dimers at and away from equilibrium separation, where we obtain mean absolute errors between
0.4 and 0.7 kcal/mol for several chemically and conformationally diverse datasets representative of
non-covalent interactions in biologically relevant molecules. We further focus on hydrogen-bonded
complexes—essential but challenging due to their directional nature—where datasets of DNA base
pairs and amino acids yield an extremely encouraging 1.4 kcal/mol error. Finally, and as a first look,
we consider IPML for denser systems: water clusters, supramolecular host-guest complexes, and the
benzene crystal.
Disciplines :
Physics
Author, co-author :
Bereau, Tristan
Distasio Jr., Robert A.
Tkatchenko, Alexandre ; University of Luxembourg > Faculty of Science, Technology and Communication (FSTC) > Physics and Materials Science Research Unit
von Lilienfeld, Anatole
External co-authors :
yes
Language :
English
Title :
Non-covalent interactions across organic and biological subsets of chemical space: Physics-based potentials parametrized from machine learning
Publication date :
15 March 2018
Journal title :
Journal of Chemical Physics
ISSN :
1089-7690
Publisher :
American Institute of Physics, New York, United States - New York
Volume :
148
Pages :
241706
Peer reviewed :
Peer Reviewed verified by ORBi
Focus Area :
Physics and Materials Science Computational Sciences