Pharmacophore-Conditioned Diffusion Model for Ligand-Based De Novo Drug Design

2025-05-15

AI Summary

For novel drug targets lacking structural or functional data, generating bioactive molecules remains challenging. This work introduces the first pharmacophore-guided 3D molecular diffusion generative model: it embeds atom-level 3D pharmacophore features into a geometric diffusion process and employs a Transformer to jointly model atomic types, 3D coordinates, and bond orders, enabling end-to-end, pharmacophore-constrained molecular graph generation without requiring protein structures. Our method achieves the first deep integration of pharmacophore features with 3D diffusion modeling, balancing ligand chemical validity and docking activity. Experiments demonstrate significant improvements over existing ligand generation methods in both pharmacophore matching accuracy and multi-target docking scores, validating its effectiveness in rational drug design and its generalizability across diverse targets.

Technology Category

• Machine Learning: Deep Generative Models & Autoencoders
• Computer Vision: Diffusion Models for Vision
• Natural Language Processing: Generation

Application Category

• Graph Algorithms and Modeling for the Web: Foundation models and LLMs for Web-related graphs
• Semantics and Knowledge: Methods to enhance, augment, integrate or synergize semantic models such as knowledge graphs and LLMs
• User Modeling, Personalization and Recommendation: Federated recommendation systems and personalization

Abstract

Developing bioactive molecules remains a central, time- and cost-heavy challenge in drug discovery, particularly for novel targets lacking structural or functional data. Pharmacophore modeling presents an alternative for capturing the key features required for molecular bioactivity against a biological target. In this work, we present PharmaDiff, a pharmacophore-conditioned diffusion model for 3D molecular generation. PharmaDiff employs a transformer-based architecture to integrate an atom-based representation of the 3D pharmacophore into the generative process, enabling the precise generation of 3D molecular graphs that align with predefined pharmacophore hypotheses. Through comprehensive testing, PharmaDiff demonstrates superior performance in matching 3D pharmacophore constraints compared to ligand-based drug design methods. Additionally, it achieves higher docking scores across a range of proteins in structure-based drug design, without the need for target protein structures. By integrating pharmacophore modeling with 3D generative techniques, PharmaDiff offers a powerful and flexible framework for rational drug design.

Problem

• Developing bioactive molecules for novel drug targets without structural data
• Generating 3D molecular graphs matching pharmacophore hypotheses precisely
• Improving docking scores without requiring target protein structures

Innovation

• Pharmacophore-conditioned diffusion model for 3D molecular generation
• Transformer-based architecture integrates 3D pharmacophore representation
• Generates molecular graphs aligning with pharmacophore hypotheses