Tag
MLIPDiff 引入一种推理时引导方法，用机器学习原子间势（UMA）的物理信息指导预训练扩散模型进行基于结构的药物设计，在不重新训练的情况下完全消除蛋白-配体碰撞并提升结合亲和力预测。
This paper introduces Sesame, a diffusion-based molecular generation model that conditions on partial molecular structure and protein pocket via spatial density maps, enabling both de novo generation and fragment-conditioned lead optimization for drug design.
This paper proposes CPES, a curvature-informed potential energy surface graph neural network for protein-ligand binding affinity prediction. It integrates physics-informed curvature representations to model conformational flexibility and achieves improved predictive performance on benchmark datasets.
Introduces InteractBind, a large-scale dataset and benchmark for fine-grained evaluation of protein-ligand models, focusing on binding-site localization and non-covalent interaction prediction. Evaluates eight existing models and finds limited binding-site localization despite strong binary binding prediction.