Advancing density functional tight-binding method for large organic molecules through equivariant neural networks

Physical Chemistry Chemical Physics, vol. 28, pp. 3949–3962

Abstract

potentials, replacing the standard pairwise DFTB repulsive potential. This advancement extends the applicability of our ML-corrected DFTB approach to larger molecules and non-covalent systems (including only C, N, O, and H atoms), going beyond the chemical space represented in the training QM datasets. The enhanced performance of EquiDTB over the standard TB methods is demonstrated by the accurate computation of the atomic forces of S66x8 molecular dimers, as well as their interaction energies. Moreover, EquiDTB can be effectively employed to explore the potential energy surfaces of large and flexible drug-like molecules-for example, to determine the minimum energy path between isomers, analyze structural transitions during dynamical simulations, compute vibrational modes, and investigate energetic rankings. The performance for single molecules slightly decreases when the DFTB electronic energy is reduced to first-order but remains superior to standard TB methods. Our work thus demonstrates that an optimal integration of an equivariant NN with QM datasets can advance the DFTB method while maintaining high efficiency, paving the way for reliable (bio)molecular simulations.

Authors 7

1. Leonardo Rafael Medrano Sandonas

   Technische Universität Dresden · Max Bergmann Zentrum für Biomaterialien

   Affiliation as printed

   Institute for Materials Science and Max Bergmann Center of Biomaterials, TUD Dresden University of Technology, 01062 Dresden, Germany

2. Mirela Puleva

   University of Luxembourg

   Affiliation as printed

   Department of Physics and Materials Science, University of Luxembourg, L-1511 Luxembourg City, Luxembourg

   Institute for Advanced Studies, University of Luxembourg, Campus Belval, L-4365 Esch-sur-Alzette, Luxembourg

3. Zekiye Erarslan

   Technische Universität Dresden · Max Bergmann Zentrum für Biomaterialien

   Affiliation as printed

   Institute for Materials Science and Max Bergmann Center of Biomaterials, TUD Dresden University of Technology, 01062 Dresden, Germany

4. Ricardo Parra Payano

   National University of Engineering

   Affiliation as printed

   Universidad Nacional de Ingeniería, Av. Túpac Amaru 210, Rímac, Lima 15333, Peru

5. Martin Stöhr

   Stanford University · SLAC National Accelerator Laboratory · University of Luxembourg

   Affiliation as printed

   Department of Chemistry and The PULSE Institute, Stanford University, Stanford, CA-94305, USA

   Department of Physics and Materials Science, University of Luxembourg, L-1511 Luxembourg City, Luxembourg

   SLAC National Accelerator Laboratory, Menlo Park CA-94025, USA

6. Gianaurelio Cuniberti Aachen Cluster of Excellence CARE

   RWTH Aachen University · Technische Universität Dresden · Climate-Neutral and Resource-Efficient Construction · Centre for Tactile Internet with Human-in-the-Loop · Max Bergmann Zentrum für Biomaterialien

   Affiliation as printed

   Cluster of Excellence CARE, TU Dresden and RWTH Aachen, Germany

   Cluster of Excellence CeTI, TU Dresden, Germany

   Institute for Materials Science and Max Bergmann Center of Biomaterials, TUD Dresden University of Technology, 01062 Dresden, Germany

7. Alexandre Tkatchenko

   University of Luxembourg

   Affiliation as printed

   Department of Physics and Materials Science, University of Luxembourg, L-1511 Luxembourg City, Luxembourg

   Institute for Advanced Studies, University of Luxembourg, Campus Belval, L-4365 Esch-sur-Alzette, Luxembourg