Abstract

Prediction of protein-ligand interactions is a critical step during the initial phase of drug discovery. We propose a novel deep-learning-based prediction model based on a graph convolutional neural network, named GraphBAR, for protein-ligand binding affinity. Graph convolutional neural networks reduce the computational time and resources that are normally required by the traditional convolutional neural network models. In this technique, the structure of a protein-ligand complex is represented as a graph of multiple adjacency matrices whose entries are affected by distances, and a feature matrix that describes the molecular properties of the atoms. We evaluated the predictive power of GraphBAR for protein-ligand binding affinities by using PDBbind datasets and proved the efficiency of the graph convolution. Given the computational efficiency of graph convolutional neural networks, we also performed data augmentation to improve the model performance. We found that data augmentation with docking simulation data could improve the prediction accuracy although the improvement seems not to be significant. The high prediction performance and speed of GraphBAR suggest that such networks can serve as valuable tools in drug discovery.

Highlights

  • Computational methods, which have been developed and improved in the past decades, have gained increased influence on drug discovery

  • It appears that this removal caused the prediction accuracy of Pafnucy slightly lower than the reported accuracy (RMSE 1.42, R 0.78), since the model showed a similar performance to the reported one when we used all the structures

  • The performance of the model was improved for datasets 2, 3, and 4, indicating that data number is critical for the performance of the model

Read more

Summary

Introduction

Computational methods, which have been developed and improved in the past decades, have gained increased influence on drug discovery. During the initial phase of the drug discovery process, the prediction of the binding affinity of a candidate ligand for a therapeutic target is an important step. Two main computational approaches, namely ligand-based and structurebased methods, have been proposed to predict protein-ligand binding affinities. Since the structural information of the target protein is not required for the prediction, the ligand-based method can be used when the biological and chemical information of the target protein and its ligands is available. This approach often utilizes molecular fingerprints as the representation of ligand structures, allowing the model to use a fixed-length vector effectively.

Methods
Results
Conclusion
Full Text
Published version (Free)

Talk to us

Join us for a 30 min session where you can share your feedback and ask us any queries you have

Schedule a call