Molecular dynamics simulations provide a mechanistic description of molecules by relying on empirical potentials. The quality and transferability of such potentials can be improved leveraging data-driven models derived with machine learning approaches. Here, we present TorchMD, a framework for molecular simulations with mixed classical and machine learning potentials. All force computations including bond, angle, dihedral, Lennard-Jones, and Coulomb interactions are expressed as PyTorch arrays and operations. Moreover, TorchMD enables learning and simulating neural network potentials. We validate it using standard Amber all-atom simulations, learning an ab initio potential, performing an end-to-end training, and finally learning and simulating a coarse-grained model for protein folding. We believe that TorchMD provides a useful tool set to support molecular simulations of machine learning potentials. Code and data are freely available at github.com/torchmd.

A generalized understanding of protein dynamics is an unsolved scientific problem, the solution of which is critical to the interpretation of the structure-function relationships that govern essential biological processes. Here, we approach this problem by constructing coarse-grained molecular potentials based on artificial neural networks and grounded in statistical mechanics. For training, we build a unique dataset of unbiased all-atom molecular dynamics simulations of approximately 9 ms for twelve different proteins with multiple secondary structure arrangements. The coarse-grained models are capable of accelerating the dynamics by more than three orders of magnitude while preserving the thermodynamics of the systems. Coarse-grained simulations identify relevant structural states in the ensemble with comparable energetics to the all-atom systems. Furthermore, we show that a single coarse-grained potential can integrate all twelve proteins and can capture experimental structural features of mutated proteins. These results indicate that machine learning coarse-grained potentials could provide a feasible approach to simulate and understand protein dynamics.

Allogeneic hematopoietic stem cell transplantation (SCT) is the only curative option for patients with primary myelofibrosis (PMF), but information on its net advantage over conventional therapies is lacking. Using ad hoc statistical analysis, we determined outcomes in 438 patients <65 years old at diagnosis who received allogenic SCT (n = 190) or conventional therapies (n = 248). Among patients at low risk per the Dynamic International Prognostic Scoring System (DIPSS) model, the relative risk of death after allogenic SCT vs those treated with nontransplant modalities was 5.6 (95% CI, 1.7-19; P = .0051); for intermediate-1 risk it was 1.6 (95% CI, 0.79-3.2; P = .19), for intermediate-2 risk, 0.55 (95% CI, 0.36-0.83; P = .005), and for high risk, 0.37 (95% CI, 0.21-0.66; P = .0007). Thus, patients with intermediate-2 or high-risk PMF clearly benefit from allogenic SCT. Patients at low risk should receive nontransplant therapy, whereas individual counseling is indicated for patients at intermediate-1 risk.