Drug-Target Interaction Prediction via Dual Laplacian Graph Regularized Matrix Completion

Area of Science:

• Bioinformatics
• Computational Biology
• Drug Discovery

Background:

• Experimental determination of drug-target interactions is costly and time-consuming.
• Accurate prediction of drug-target interactions is crucial for efficient drug discovery and development.
• Computational methods are needed to accelerate the identification of potential drug-target interactions.

Purpose of the Study:

• To propose an effective computational model for predicting unknown drug-target interactions.
• To develop a novel approach that integrates drug chemical structure and target genomic sequence similarities.
• To improve the accuracy and efficiency of drug-target interaction prediction.

Main Methods:

• Developed a dual Laplacian graph regularized matrix completion (DLGRMC) model.
• Transformed drug-target interaction prediction into a matrix completion problem.
• Utilized drug chemical structure and target genomic sequence similarities via dual Laplacian graph regularization.
• Employed an indicator matrix to preserve experimentally confirmed interactions.
• Solved the constrained matrix completion using an iterative strategy based on the Augmented Lagrange Multiplier algorithm.

Main Results:

• DLGRMC demonstrated superior performance compared to state-of-the-art methods on five benchmark datasets.
• Evaluations based on 10-fold cross-validation showed improved AUPR values and PR curves.
• Case studies confirmed DLGRMC's ability to successfully predict experimentally validated drug-target interactions.

Conclusions:

• DLGRMC is an effective computational tool for predicting drug-target interactions.
• The model successfully integrates diverse biological data for enhanced prediction accuracy.
• DLGRMC offers a promising approach to accelerate drug discovery and development.