Pharmacophore-Conditioned Diffusion Model for Ligand-Based De Novo Drug Design
- 1Carnegie Mellon University , , , ,
- 2AbbVie Bioresearch Center , , ,
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.
Related
- GCMDiff: a condition-guided diffusion framework for pharmacophore-fragment drug discovery and target-specific molecular design
- FLOWR: flow matching for structure-aware de novo, interaction- and fragment-based ligand generation
- A Scoping Review of Diffusion-Based De Novo Molecular Generation: Architectures, Evaluation, and Open Challenges for Computational Drug Discovery
- De novo Design of a Vanadium Haloperoxidase
- DEG2MOL: Conditional Latent Flow Matching for Transcriptome-Guided de novo Drug Design
BibTeX
@article{alakhdar2026pharmacophoreconditioned,
title = {{Pharmacophore-Conditioned Diffusion Model for Ligand-Based De Novo Drug Design}},
author = {Amira A. Alakhdar and Barnabás Póczos and Newell R. Washburn and Kevin Cusack},
journal = {Journal of Chemical Information and Modeling},
year = {2026},
month = oct,
doi = {10.1021/acs.jcim.6c01046},
url = {https://doi.org/10.1021/acs.jcim.6c01046},
note = {CC BY},
}
RIS
TY - JOUR
TI - Pharmacophore-Conditioned Diffusion Model for Ligand-Based De Novo Drug Design
AU - Amira A. Alakhdar
AU - Barnabás Póczos
AU - Newell R. Washburn
AU - Kevin Cusack
PY - 2026
DA - 2026/10/27
JO - Journal of Chemical Information and Modeling
DO - 10.1021/acs.jcim.6c01046
UR - https://doi.org/10.1021/acs.jcim.6c01046
AB - 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.
SN - 1549-9596
SN - 1549-960X
ER -