Abstract:The design of drug molecules is a critical stage in the drug discovery process. The use of pharmacophore models in structure-based drug discovery has long played an important role in efficient development. Significant progress has been made in recent years in the generation of 3D molecules via deep generation models. However, while many existing models have succeeded in incorporating structural information of target proteins, they have not been able to address important interactions between proteins and drug molecules, especially hydrogen bonds, explicitly. In this study, we propose DiffInt as a novel structure-based approach that explicitly addresses interactions. The model naturally incorporates hydrogen bonds between the protein and ligand by treating them as pseudoparticles. The experimental results show that DiffInt reproduces hydrogen bonds and that the hydrogen binding energies significantly outperform those of existing models. To facilitate the use of our tool for generating new drug molecules based on any protein three-dimensional structure, we have made the source code and trained model available on GitHub (https://github.com/sekijima-lab/DiffInt) under the MIT license, with the execution environment provided on Google Colab.

What problem does this paper attempt to address?

The main objective of this paper is to propose a new method to address key issues in structure-based drug design, particularly the explicit modeling of hydrogen bond interactions between drug molecules and target proteins. Specifically, the study aims to address the following issues: 1. **Explicitly handling hydrogen bond interactions between protein and ligand**: Existing generative models, while considering the structural information of the target protein, fall short in explicitly handling the critical interactions (especially hydrogen bonds) between proteins and drug molecules. 2. **Improving binding affinity**: By considering specific hydrogen bond interactions during the generation process, the goal is to generate new drug molecules with higher binding affinity. 3. **Overcoming limitations of existing models**: Addressing issues in sequence-based generative methods, such as cumulative errors leading to unrealistic molecular structures, and the limitations of existing diffusion models that fail to adequately consider pharmacophore models. To solve the above problems, the researchers proposed a new method called DiffInt. DiffInt is a diffusion model based on the pharmacophore model, with a particular focus on hydrogen bond interactions between proteins and ligands. This method introduces interactive particles to simulate hydrogen bonds, thereby enabling the explicit reconstruction of these interactions during the generation process. Experimental results show that compared to benchmark models, DiffInt performs better in reconstructing hydrogen bonds, hydrogen bond binding energy, and high affinity, demonstrating better drug properties and higher diversity in generated molecules. In summary, this study aims to significantly enhance the binding affinity and pharmacological properties of generated molecules by improving the handling of hydrogen bonds in the molecular generation process, thereby accelerating the process of new drug discovery.

DiffInt: A Pharmacophore-Aware Diffusion Model for Structure-Based Drug Design with Explicit Hydrogen Bond Interaction Guidance

A Unified Conditional Diffusion Framework for Dual Protein Targets Based Bioactive Molecule Generation

Interaction-based Retrieval-augmented Diffusion Models for Protein-specific 3D Molecule Generation

AUTODIFF: Autoregressive Diffusion Modeling for Structure-based Drug Design

DecompDiff: Diffusion Models with Decomposed Priors for Structure-Based Drug Design

Manifold-Constrained Nucleus-Level Denoising Diffusion Model for Structure-Based Drug Design

Protein-Ligand Interaction Prior for Binding-aware 3D Molecule Diffusion Models

MolSnapper: Conditioning Diffusion for Structure Based Drug Design

Binding-Adaptive Diffusion Models for Structure-Based Drug Design

PIDiff: Physics informed diffusion model for protein pocket-specific 3D molecular generation

Structure-based Drug Design with Equivariant Diffusion Models

Aligning Target-Aware Molecule Diffusion Models with Exact Energy Optimization

Complex-based Ligand-Binding Proteins Redesign by Equivariant Diffusion-based Generative Models

QDπ: A Quantum Deep Potential Interaction Model for Drug Discovery

DrugDiff - small molecule diffusion model with flexible guidance towards molecular properties

3D Equivariant Diffusion for Target-Aware Molecule Generation and Affinity Prediction

DiffBP: Generative Diffusion of 3D Molecules for Target Protein Binding

Diff-Shape: A Novel Constrained Diffusion Model for Shape based De Novo Drug Design

DiffBindFR: An SE(3) Equivariant Network for Flexible Protein-Ligand Docking