Abstract:The flexibility of the receptor and ligand is vital in understanding the process of drug binding. To exploit this dynamic information, a spatial‐temporal pre‐training protocol, dubbed ProtMD, is proposed. The algorithm offers a perspective of pre‐training to learn the flexibility of conformations with molecular dynamics simulations. This work could provide computer‐aided guidance in drug design and has the potential to boost pre‐clinical drug development. The latest biological findings observe that the motionless "lock‐and‐key" theory is not generally applicable and that changes in atomic sites and binding pose can provide important information for understanding drug binding. However, the computational expenditure limits the growth of protein trajectory‐related studies, thus hindering the possibility of supervised learning. A spatial‐temporal pre‐training method based on the modified equivariant graph matching networks, dubbed ProtMD which has two specially designed self‐supervised learning tasks: atom‐level prompt‐based denoising generative task and conformation‐level snapshot ordering task to seize the flexibility information inside molecular dynamics (MD) trajectories with very fine temporal resolutions is presented. The ProtMD can grant the encoder network the capacity to capture the time‐dependent geometric mobility of conformations along MD trajectories. Two downstream tasks are chosen to verify the effectiveness of ProtMD through linear detection and task‐specific fine‐tuning. A huge improvement from current state‐of‐the‐art methods, with a decrease of 4.3% in root mean square error for the binding affinity problem and an average increase of 13.8% in the area under receiver operating characteristic curve and the area under the precision‐recall curve for the ligand efficacy problem is observed. The results demonstrate a strong correlation between the magnitude of conformation's motion in the 3D space and the strength with which the ligand binds with its receptor.

Pretrainable Geometric Graph Neural Network for Antibody Affinity Maturation

Pretrainable geometric graph neural network for antibody affinity maturation

Deep Geometric Framework to Predict Antibody-Antigen Binding Affinity

Sequence-based deep learning antibody design for in silico antibody affinity maturation

Pre‐Training of Equivariant Graph Matching Networks with Conformation Flexibility for Drug Binding

GAABind: a Geometry-Aware Attention-Based Network for Accurate Protein-Ligand Binding Pose and Binding Affinity Prediction

Pre-training Antibody Language Models for Antigen-Specific Computational Antibody Design

Conditional Antibody Design as 3D Equivariant Graph Translation

AlphaBind, a Domain-Specific Model to Predict and Optimize Antibody-Antigen Binding Affinity

Pre-training of Graph Neural Network for Modeling Effects of Mutations on Protein-Protein Binding Affinity

GeoAB: Towards Realistic Antibody Design and Reliable Affinity Maturation

SAGERank: Inductive Learning of Protein-Protein Interaction from Antibody-Antigen Recognition Using Graph Sample and Aggregate Networks Framework

GEMS: A Generalizable GNN Framework For Protein-Ligand Binding Affinity Prediction Through Robust Data Filtering and Language Model Integration

Incorporating Pre-training Paradigm for Antibody Sequence-Structure Co-design

ProAffinity-GNN: A Novel Approach to Structure-based Protein-Protein Binding Affinity Prediction via a Curated Dataset and Graph Neural Networks

Deep Geometric Representations for Modeling Effects of Mutations on Protein-Protein Binding Affinity.

Multi-task bioassay pre-training for protein-ligand binding affinity prediction

GIANT: Protein-Ligand Binding Affinity Prediction via Geometry-aware Interactive Graph Neural Network

DeepBindGCN: Integrating Molecular Vector Representation with Graph Convolutional Neural Networks for Protein–Ligand Interaction Prediction

A Hierarchical Training Paradigm for Antibody Structure-sequence Co-design