Coarse graining enables the investigation of molecular dynamics for larger systems and at longer timescales than is possible at an atomic resolution. However, a coarse graining model must be formulated such that the conclusions we draw from it are consistent with the conclusions we would draw from a model at a finer level of detail. It has been proved that a force matching scheme defines a thermodynamically consistent coarse-grained model for an atomistic system in the variational limit. Wang et al. [ACS Cent. Sci. 5, 755 (2019)] demonstrated that the existence of such a variational limit enables the use of a supervised machine learning framework to generate a coarse-grained force field, which can then be used for simulation in the coarse-grained space. Their framework, however, requires the manual input of molecular features to machine learn the force field. In the present contribution, we build upon the advance of Wang et al. and introduce a hybrid architecture for the machine learning of coarse-grained force fields that learn their own features via a subnetwork that leverages continuous filter convolutions on a graph neural network architecture. We demonstrate that this framework succeeds at reproducing the thermodynamics for small biomolecular systems. Since the learned molecular representations are inherently transferable, the architecture presented here sets the stage for the development of machine-learned, coarse-grained force fields that are transferable across molecular systems.

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 .

Along with a growing interest in mRNA-based gene therapies, efforts are increasingly focused on reaching the full translational potential of mRNA, as a major obstacle for in vivo applications is sufficient expression of exogenously delivered mRNA. One method to overcome this limitation is chemically modifying the 7-methylguanosine cap at the 5′ end of mRNA (m7Gppp-RNA). We report a novel class of cap analogs designed as reagents for mRNA modification. The analogs carry a 1,2-dithiodiphosphate moiety at various positions along a tri- or tetraphosphate bridge, and thus are termed 2S analogs. These 2S analogs have high affinities for translation initiation factor 4E, and some exhibit remarkable resistance against the SpDcp1/2 decapping complex when introduced into RNA. mRNAs capped with 2S analogs combining these two features exhibit high translation efficiency in cultured human immature dendritic cells. These properties demonstrate that 2S analogs are potentially beneficial for mRNA-based therapies such as anti-cancer immunization.

The predominant view in structure-based drug design is that small-molecule ligands, once bound to their target structures, display a well-defined binding mode. However, structural stability (robustness) is not necessary for thermodynamic stability (binding affinity). In fact, it entails an entropic penalty that counters complex formation. Surprisingly, little is known about the causes, consequences and real degree of robustness of protein-ligand complexes. Since hydrogen bonds have been described as essential for structural stability, here we investigate 469 such interactions across two diverse structure sets, comprising of 79 druglike and 27 fragment ligands, respectively. Completely constricted protein-ligand complexes are rare and may fulfill a functional role. Most complexes balance order and disorder by combining a single anchoring point with looser regions. 25% do not contain any robust hydrogen bond and may form loose structures. Structural stability analysis reveals a hidden layer of complexity in protein-ligand complexes that should be considered in ligand design.

The accurate prediction of protein−ligand binding affinities is crucial for drug discovery. Alchemical free energy calculations have become a popular tool for this purpose. However, the accuracy and reliability of these methods can vary depending on the methodology. In this study, we evaluate the performance of a relative binding free energy protocol based on the alchemical transfer method (ATM), a novel approach based on a coordinate transformation that swaps the positions of two ligands. The results show that ATM matches the performance of more complex free energy perturbation (FEP) methods in terms of Pearson correlation but with marginally higher mean absolute errors. This study shows that the ATM method is competitive compared to more traditional methods in speed and accuracy and offers the advantage of being applicable with any potential energy function.