4 April 2023 in Chemical Science
Deep-HP is a scalable extension of the Tinker-HP multi-GPU molecular dynamics (MD) package enabling the use of Pytorch/TensorFlow Deep Neural Network (DNN) models. Deep-HP increases DNNs' MD capabilities by orders of magnitude offering access to ns simulations for 100k-atom biosystems while offering the possibility of coupling DNNs to any classical (FFs) and many-body polarizable (PFFs) force fields. It allows therefore the introduction of the ANI-2X/AMOEBA hybrid polarizable potential designed for ligand binding studies where solvent–solvent and solvent–solute interactions are computed with the AMOEBA PFF while solute–solute ones are computed by the ANI-2X DNN. ANI-2X/AMOEBA explicitly includes AMOEBA's physical long-range interactions via an efficient Particle Mesh Ewald implementation while preserving ANI-2X's solute short-range quantum mechanical accuracy. The DNN/PFF partition can be user-defined allowing for hybrid simulations to include key ingredients of biosimulation such as polarizable solvents, polarizable counter ions, etc.… ANI-2X/AMOEBA is accelerated using a multiple-timestep strategy focusing on the model's contributions to low-frequency modes of nuclear forces. It primarily evaluates AMOEBA forces while including ANI-2X ones only via correction-steps resulting in an order of magnitude acceleration over standard Velocity Verlet integration. Simulating more than 10 μs, we compute charged/uncharged ligand solvation free energies in 4 solvents, and absolute binding free energies of host–guest complexes from SAMPL challenges. ANI-2X/AMOEBA average errors are discussed in terms of statistical uncertainty and appear in the range of chemical accuracy compared to experiment. The availability of the Deep-HP computational platform opens the path towards large-scale hybrid DNN simulations, at force-field cost, in biophysics and drug discovery.
25 July 2024 in The Journal of Chemical Physics
Neural network interatomic potentials (NNPs) have recently proven to be powerful tools to accurately model complex molecular systems while bypassing the high numerical cost of ab initio molecular dynamics simulations. In recent years, numerous advances in model architectures as well as the development of hybrid models combining machine-learning (ML) with more traditional, physically motivated, force-field interactions have considerably increased the design space of ML potentials. In this paper, we present FeNNol, a new library for building, training, and running force-field-enhanced neural network potentials. It provides a flexible and modular system for building hybrid models, allowing us to easily combine state-of-the-art embeddings with ML-parameterized physical interaction terms without the need for explicit programming. Furthermore, FeNNol leverages the automatic differentiation and just-in-time compilation features of the Jax Python library to enable fast evaluation of NNPs, shrinking the performance gap between ML potentials and standard force-fields. This is demonstrated with the popular ANI-2x model reaching simulation speeds nearly on par with the AMOEBA polarizable force-field on commodity GPUs (graphics processing units). We hope that FeNNol will facilitate the development and application of new hybrid NNP architectures for a wide range of molecular simulation problems.
17 December 2025 in ChemRxiv
While artificial intelligence has revolutionized the prediction of static protein structures, characterizing their dynamics and interactions with drug candidates remains a computational bottleneck. Here, we introduce FeNNix-Bio1, a foundation machine learning model designed to power accurate, reactive atomistic simulations of biological systems at an unprecedented speed and scalability. Trained exclusively on synthetic quantum chemistry data, FeNNix-Bio1 accurately captures complex condensed-phase phenomena such as ion solvation and subtle liquid water properties for which it outperforms state-of-the-art specialized force fields. We demonstrate its versatility across a full spectrum of drug design applications, including the calculation of hydration free energies (HFEs), the reversible folding of small proteins, the simulation of protein-ligand absolute binding free energies and chemical reactions. Notably, FeNNix-Bio1 sets a new standard for the precise prediction of HFEs for the more than 600 molecules of the Freesolv dataset, providing sub-kcal/mol accuracy. By enabling scalable, quantum-accurate molecular dynamics without the need for manual parametrization, FeNNix-Bio1 bridges the gap between static structure prediction and dynamic biological reality: it is likely to have a strong impact in Drug Design.