Concept-Driven Deep Learning for Enhanced Protein-Specific Molecular Generation
This addresses protein-specific molecular generation for drug discovery, representing an incremental improvement over existing fragment-based approaches.
The paper tackles the problem of generating molecules for specific protein targets by addressing limitations in synthetic feasibility, drug-likeness, and interpretability of existing methods. Their fragment-based framework combining concept-based neural networks and diffusion models achieved a 4% increase in drug-likeness and 6% improvement in synthetic feasibility.
In recent years, deep learning techniques have made significant strides in molecular generation for specific targets, driving advancements in drug discovery. However, existing molecular generation methods present significant limitations: those operating at the atomic level often lack synthetic feasibility, drug-likeness, and interpretability, while fragment-based approaches frequently overlook comprehensive factors that influence protein-molecule interactions. To address these challenges, we propose a novel fragment-based molecular generation framework tailored for specific proteins. Our method begins by constructing a protein subpocket and molecular arm concept-based neural network, which systematically integrates interaction force information and geometric complementarity to sample molecular arms for specific protein subpockets. Subsequently, we introduce a diffusion model to generate molecular backbones that connect these arms, ensuring structural integrity and chemical diversity. Our approach significantly improves synthetic feasibility and binding affinity, with a 4% increase in drug-likeness and a 6% improvement in synthetic feasibility. Furthermore, by integrating explicit interaction data through a concept-based model, our framework enhances interpretability, offering valuable insights into the molecular design process.