Schrodinger (United States)

Machine learning interatomic potentials (MLIPs), also known as machine learning force fields (MLFFs), offer scalable means of simulating complex systems and processes at \textit{ab initio} level accuracy. One such process is the critical yet still poorly understood formation of the solid electrolyte interphase (SEI) at the anode of a Li-ion battery (LIB) during the first charge cycle, where electrochemical reduction of the electrolyte leads to the generation of decomposition products. MLIPs are uniquely poised to atomistically describe these electrochemical processes, as they are not as affected by the same limitations in bonding and electron transfer as classical force fields. Nonetheless, training MLIPs to run accurate dynamics of a condensed phase with two different oxidation states, such as in electrochemistry, is challenging for many architectures. In this work, we show that by using MPNICE, a message passing MLIP architecture with iterative charge equilibration, we are able to accurately (within 1 kcal/mol) train models along two potential energy surfaces (reduced and unreduced) for LIB-relevant electrolyte systems. Importantly, we demonstrate strategies for sampling and training to examples of anion radicals of these species, which often are not centered on any atom (off-center radicals, or OCRs). We additionally discuss well known limitations of global charge equilibration (Qeq) algorithms in erroneously de-localizing charge, and test methods to alleviate the impact on resulting dynamics. Simulations using these models reveal new insights into electrolyte reduction and considerations for the realistic simulation of electron transfer processes in the condensed phase.

Deep learning co-folding models for end-to-end protein–ligand structure prediction mark a major advance beyond AlphaFold2, yet their reliability for decision-making in drug discovery remains unclear. Here, we benchmark Boltz, a state-of-the-art co-folding model, using a curated set of ligand-bound human G protein-coupled receptors (GPCRs) from families unseen during training. We find that the receptor backbones are generally predicted with reasonable accuracy, but ligand poses often deviate significantly from experimental structures. We then evaluate physics-based refinement with rigid-receptor (Glide) and induced-fit docking (IFD-MD) methods, which recover more than half of the misplaced ligands to near-experimental accuracy. As conventional evaluations for co-folded structures focus on distance-based metrics such as root-mean-squared deviation (RMSD), which can miss subtle but consequential binding-site errors, we carry out a further assessment of Boltz performance using free-energy perturbation (FEP+), which is both accurate and sensitive to starting-structure quality, on curated congeneric ligand series with known binding affinities that target the GPCRs. A significant fraction of the 14 congeneric series tested in this fashion fail to reproduce experimental binding affinities via FEP+ when employing the Boltz generated complex, even when the binding-site RMSD is low in some cases. IFD-MD rescues these failures and restores retrospective FEP signals to native-like level for all of these series. Together, these results delineate current generalization and usability limits of co-folded GPCR–ligand complexes and motivate a workflow that pairs deep learning predictions with physics-based refinement and validation before high-stakes decisions in drug discovery.

Disrupting protein homeostasis in Mycobacterium tuberculosis (Mtb) by targeting the ClpC1P1P2 proteolytic complex is a promising anti-TB strategy. We synthesized conformationally constrained monomeric and dimeric rufomycin 4/6 (RUF) analogs via click chemistry. While most monomeric analogs were inactive, dimeric analogs displayed potent anti-Mtb activity. Surface plasmon resonance revealed tight, slow-dissociating binding of dimers to the ClpC1 N-terminal domain (ClpC1NTD), indicating prolonged residency time. X-ray crystallography and size exclusion chromatography demonstrated that dimeric analogs induce NTD dimerization, likely shifting the equilibrium toward enzymatically active hexamers rather than inactive decamers. Dimers enhanced ATPase activity over 10-fold, surpassing that of ecumicin (8-fold) and far exceeding RUF (<2-fold). Notably, our dimers bind two NTDs per molecule (1:2), contrasting with RUF (1:1) and ecumicin (2:1), suggesting a distinct activation mechanism. These RUF-based click chemistry dimers represent potent ClpC1 modulators with extended residency and novel stoichiometry, offering promising tools for studying protein degradation.

Powered by dramatic advances in computer hardware, the advent of ultralarge make-on-demand virtual libraries, and a shift in small-molecule discovery toward more challenging targets with limited known actives, there has been a growing interest in the development of performant virtual screening methods that can reliably deliver novel hits. We report on a new method called Glide WS, that builds on our earlier efforts (WScore) to introduce an explicit representation of water structure and dynamics to an empirical scoring function suitable for high-throughput docking. This scoring function has been carefully tuned using absolute binding free energy perturbation calculations (ABFEP). Compared with Glide SP, Glide WS offers significant gains in the two primary tasks for molecular docking in drug discovery, pose prediction and virtual screening enrichment. For docking accuracy, Glide WS achieves a self-docking accuracy of 92% on a diverse set of 1477 protein ligand complexes as compared to 85% for Glide SP, using a criterion of 2.5 Å. We also demonstrate significantly improved virtual screening enrichment using a diverse data set covering of 38 targets together with three different computationally generated libraries of decoys, combined with standard known ChEMBL actives. We focus on ligands ranked in the top few percent of the database (the subset that is relevant to practical virtual screening efforts) and demonstrate that, along with improved enrichment of ChEMBL actives, Glide WS achieves a remarkable reduction in the number of poorly scoring decoys (as calibrated by ABFEP calculations), across a high percentage of targets, as compared to Glide SP. These results suggest that considerably higher hit rates will be observed, as compared to conventional rigid receptor docking, in practical virtual screening applications.

Cryo-electron microscopy can now routinely deliver atomic resolution structures for a variety of biological systems. The relevance and value of these structures are directly related to their ability to help rationalize experimental observables, which in turn depend on the quality of the model built into the density map. Coupling traditional model-building tools with physics-based methods, such as docking, simulation, and modern force fields, has been shown to improve the quality of the resulting structures. Here, we survey the landscape of these hybrid approaches, highlighting their usefulness for medium- and low-resolution datasets, as well as for structures of small molecules, and make the argument that the community stands to benefit from their inclusion in model building and refinement workflows.

Powered by dramatic advances in computer hardware, the advent of ultra-large make-on-demand virtual libraries, and a shift in small molecule discovery toward more challenging targets with limited known actives, there has been a growing interest in the development of performant virtual screening methods that can reliably deliver novel hits. We report on a new method called Glide WS, that builds on our earlier efforts (WScore) to introduce an explicit representation of water structure and dynamics to an empirical scoring function suitable for high-throughput docking. This scoring function has been carefully tuned using Absolute Binding Free Energy Perturbation calculations (ABFEP). Compared with Glide SP, Glide WS offers significant gains in the two primary tasks for molecular docking in drug discovery, pose prediction and virtual screening enrichment. For docking accuracy, Glide WS achieves a self-docking accuracy of 92% on a diverse set of 1477 protein ligand complexes as compared to only 85% for Glide SP, using a criterion of 2.5 Å. We also demonstrate significantly improved virtual screening enrichment using a diverse dataset covering of 38 targets together with three different computationally generated libraries of decoys, combined with standard known ChEMBL actives). We focus ligands ranked in the top few percent of the database (the subset that is relevant to practical virtual screening efforts) and demonstrate that, along with improved enrichment of ChEMBL actives, Glide WS achieves a remarkable reduction in the number of poorly scoring decoys (as calibrated by ABFEP calculations), across a high percentage of targets, as compared to Glide SP. These results suggest that considerably higher hit rates will be observed, as compared to conventional rigid receptor docking, in practical virtual screening applications.