Generating Developable 3D Molecules via Pocket-Conditioned Diffusion and Property-Aware Optimization

2026-07-14

AI Summary

This work addresses the challenge in structure-based drug design where existing diffusion models struggle to simultaneously optimize target binding affinity and molecular developability, particularly ADMET properties. The authors propose conDitar-dev, a novel framework that decouples pocket and ligand representations and integrates multi-scale pocket encoding (msPRL), pocket-conditioned diffusion generation (conDitar), and a property-aware optimization strategy (paOPT) to enable atomic- and molecular-level co-modeling with real-time ADMET optimization during generation. Evaluated on human disease targets, the method achieves an average binding energy of −8.85 kcal/mol and improves ADMET performance by up to 73%. Experimentally validated molecules targeting PD-L1 and CSF1R demonstrate micromolar binding affinity and nanomolar inhibitory activity.

Abstract

Drug discovery and development is time-consuming and resource-intensive, motivating computational approaches such as diffusion models for de novo drug design. Many such models follow the structure-based drug design (SBDD) paradigm, generating molecules to fit a target binding pocket. However, existing diffusion-based SBDD methods typically couple pocket and ligand representation learning, model interactions only at the atom level, and prioritize binding affinity over other developability properties. Here, we introduce conDitar-dev, a conditional diffusion-based SBDD framework for generating ligands with strong binding affinities and favorable ADMET properties. It consists of three modules: msPRL, a pretrained multi-scale pocket representation learning module; conDitar, a pocket-conditioned diffusion model guided by msPRL representations; and paOPT, a generation-time method for optimizing ligand developability. On a newly curated benchmark of human disease targets, conDitar outperforms state-of-the-art SBDD baselines, achieving an average binding score of -8.85 kcal/mol. Across five ADMET properties, conDitar-dev improves performance by up to 73% over conDitar. To further validate the abilities of conDitar-dev to generate developable molecules, we have applied it to two validated druggable targets: programmed death-ligand 1 (PD-L1) and colony-stimulating factor 1 receptor (CSF1R) proteins. Top-ranked generatively designed molecules and their analogs have been experimentally synthesized and biologically tested. Two molecules generated directly by conDitar-dev for PD-L1 exhibited SPR-derived $K_D$ values of 3.49 and 3.75 $μ$M, respectively. Hit expansion based on conDitar-dev-designed molecules identified selective CSF1R inhibitors with IC$_{50}$ values as low as 200 nM, while also uncovering opportunities for drug repositioning.

Problem

• drug developability
• structure-based drug design
• ADMET properties
• binding affinity
• molecular generation

Innovation

• pocket-conditioned diffusion
• multi-scale pocket representation
• property-aware optimization
• developable molecule generation
• structure-based drug design