Kai Zhu 「祝凯」

Rare events govern many important molecular processes but remain difficult to characterize within accessible simulation timescales. This challenge has motivated machine-learning methods to construct low-dimensional collective variables for specific objectives, including state discrimination, slow-mode identification, and committor-function approximation. Here we introduce SelfTICA, a self-supervised framework that uses contrastive learning on time-lagged configurations to learn a latent representation of the slow modes governing relevant transitions. Once learned, this representation is frozen and reused across downstream tasks, including collective variables construction, enhanced sampling, free-energy estimation, and committor learning. Compared with direct slow-mode optimization, SelfTICA improves training stability and provides collective variables even from limited and exploratory trajectories, accelerating rare-event sampling and enabling accurate free-energy estimation. The dynamical information encoded during pretraining also accelerates committor learning and enables characterization of transition-state regions. These results show that contrastive learning of slow dynamical representations provides a common foundation for rare-event sampling and mechanistic analysis.

Intrinsically disordered proteins are challenging drug targets because they lack stable three-dimensional structures and instead exist as dynamic conformational ensembles. In this work, we developed a machine learning-based enhanced sampling workflow to target the intrinsically disordered N-terminal domain of the androgen receptor (AR-NTD), a therapeutic target for drug-resistant prostate cancer. By combining enhanced sampling simulations with machine-learning collective variables, we characterized the conformational landscape of the Tau-5 region, identified druggable conformations, and revealed key molecular features involved in ligand recognition. These insights enabled structure-based virtual screening and led to the identification of K53, an AR-NTD antagonist with anti-proliferative activity in enzalutamide-resistant prostate cancer cells.

Molecular dynamics simulations hold great promise for providing insight into the microscopic behavior of complex molecular systems. However, their effectiveness is often constrained by long timescales associated with rare events. Enhanced sampling methods have been developed to address these challenges, and recent years have seen a growing integration with machine learning techniques. This Review provides a comprehensive overview of how they are reshaping the field, with a particular focus on the data-driven construction of collective variables. Furthermore, these techniques have also improved biasing schemes and unlocked novel strategies via reinforcement learning and generative approaches. In addition to methodological advances, we highlight applications spanning different areas, such as biomolecular processes, ligand binding, catalytic reactions, and phase transitions. We conclude by outlining future directions aimed at enabling more automated strategies for rare-event sampling.

Accurate atomistic biomolecular simulations are vital for disease mechanism understanding, drug discovery, and biomaterial design, but existing simulation methods exhibit significant limitations. Classical force fields are efficient but lack accuracy for transition states and fine conformational details critical in many chemical and biological processes. Quantum Mechanics methods are highly accurate but computationally infeasible for large-scale or long-time simulations. AI-based force fields aim to achieve quantum-level accuracy with efficiency but struggle to balance many-body modeling complexity, accuracy, and speed. To overcome these challenges, we introduce LiTEN, an equivariant neural network with Tensorized Quadrangle Attention. TQA efficiently models three- and four-body interactions with linear complexity by reparameterizing high-order tensor features via vector operations. Building on LiTEN, LiTEN-FF is a robust AI force-field foundation model pre-trained on the nablaDFT dataset and fine-tuned on SPICE. LiTEN achieves strong performance across rMD17, MD22, and Chignolin while enabling downstream biomolecular modeling tasks including conformer searches, geometry optimization, and free-energy-surface construction.