Xiao Huang

Electron bifurcation reactions separate electron pairs into high and low potential pools, and these reactions are central to the bioenergetics of living systems. Here, we used kinetic analysis and machine learning to analyze a diverse set of structural and electrochemical landscapes that may guide the design of molecular architectures that could serve as experimental targets that would function to bifurcate holes using light. We find that strong electrostatic repulsion between the holes enhances the quantum yield for bifurcation but reduces the energy efficiency of the process. We find that the quantum yield for hole bifurcation is enhanced by positioning the hot-hole pathway cofactor farther from the hole bifurcation site than its cold-hole pathway counterpart. This integrated design and optimization approach provides design strategies for de novo structures that could realize light-drive hole bifurcation, advancing the aim of employing bioinspired electron bifurcation for energy conversion, photocatalysis, and electrocatalysis. Beyond the specific light-driven hole-bifurcation architecture, our combined kinetic–statistical–machine-learning approach is transferable to other multi-particle, multi-site charge-transport network design challenges, opening paths for designing photochemical and catalytic networks, as well as for designing functional redox networks.

We report a reinforcement-learning-enhanced generative chemistry pipeline for the de novo design of small-molecule inhibitors targeting Epidermal Growth Factor Receptor (EGFR). Starting from a pretrained ChemBERTa language model fine-tuned on high-affinity EGFR in- hibitors, we introduce a multicomponent reward function combining predicted potency (40%), drug-likeness (QED, 25%), synthetic accessibility (SA, 15%), and novelty relative to the train- ing library (20%). Through policy-gradient optimization over 500 iterations, the model learns to produce chemically valid, diverse, and novel scaffolds enriched for high composite rewards. Compared with the prior policy, the RL-tuned generator achieves a 20 % increase in mean re- ward and yields a threefold expansion in unique Bemis-Murcko cores. High-throughput dock- ing against the EGFR kinase domain (PDB ID: 1M17) demonstrates that the newly generated library attains a median predicted binding affinity of –9.2 kcal/mol, significantly surpassing the –8.5 kcal/mol baseline of known inhibitors. An exemplar generated ligand recapitulates key hinge-binding interactions while presenting a novel solvent-exposed substituent for further optimization. This study illustrates the power of integrating language-model pretraining with reinforcement learning and composite reward engineering to accelerate target-focused drug discovery.

Electron bifurcation reactions divide electrons from two-electron donors into high- and low-energy pools by transporting charge on spatially separated low- and high-potential electron hopping pathways. Bifurcation delivers electrons at potentials that drive downstream reactions in photosynthesis, respiration, and biocatalysis. Recent theoretical studies have described the requirements for effective ground-state electron bifurcation. The aim of this study is to design synthetic bifurcation constructs that can be driven by light. We describe a strategy to bifurcate holes (oxidizing equivalents) efficiently with light, and we present an illustrative energy landscape that could support this design. The design focuses on the electrochemical potentials and distances between cofactors. The analysis finds that hole bifurcation may be driven efficiently with light, guiding the further development of bioinspired networks that bifurcate charge and deliver the charges with prescribed electrochemical potentials.

Electron bifurcation reactions divide electrons from two-electron donors into high- and low-energy pools by transporting charge on spatially separated low- and high-potential electron hopping pathways. Bifurcation delivers electrons at potentials that drive downstream reactions in photosynthesis, respiration, and biocatalysis. Recent theoretical studies have described the requirements for effective ground-state electron bifurcation. The aim of this study is to design synthetic bifurcation constructs that can be driven by light. We describe a strategy to bifurcate holes (oxidizing equivalents) efficiently with light, and we present an illustrative energy landscape that could support this design. The design focuses on the electrochemical potentials and distances between cofactors. The analysis finds that hole bifurcation may be driven efficiently with light, guiding the further development of bioinspired networks that bifurcate charge and deliver the charges with prescribed electrochemical potentials.

The availability of large, high-quality datasets is crucial for artificial intelligence design and discovery in chemistry. Despite the essential roles of solvents in chemistry, the rapid computational dataset generation of solution-phase molecular properties at the quantum mechanical level of theory was previously hampered by the complicated simulation procedure. Software toolkits that can automate the procedure to set up high-throughput explicit-solvent quantum chemistry (QC) calculations for arbitrary solutes and solvents in an open-source framework are still lacking. We developed AutoSolvate, an open-source toolkit, to streamline the workflow for QC calculation of explicitly solvated molecules. It automates the solvated-structure generation, force field fitting, configuration sampling, and the final extraction of microsolvated cluster structures that QC packages can readily use to predict molecular properties of interest. AutoSolvate is available through both a command line interface and a graphical user interface, making it accessible to the broader scientific community. To improve the quality of the initial structures generated by AutoSolvate, we investigated the dependence of solute–solvent closeness on solute/solvent identities and trained a machine learning model to predict the closeness and guide initial structure generation. Finally, we tested the capability of AutoSolvate for rapid dataset curation by calculating the outer-sphere reorganization energy of a large dataset of 166 redox couples, which demonstrated the promise of the AutoSolvate package for chemical discovery efforts.