NMTF-DTI: A Nonnegative Matrix Tri-factorization Approach With Multiple Kernel Fusion for Drug-Target Interaction...

Area of Science:

• Bioinformatics
• Computational Chemistry
• Drug Discovery

Background:

• Drug-target interactions (DTIs) are crucial for drug development but experimentally determined through costly and time-consuming processes.
• Computational methods offer a cost-effective alternative to predict DTIs, reducing risks in experimental drug discovery.

Purpose of the Study:

• To propose an effective computational approach, NMTF-DTI, for predicting drug-target interaction scores.
• To enhance DTI prediction accuracy by utilizing multiple similarity kernels and Laplacian regularization.

Main Methods:

• Nonnegative matrix tri-factorization (NMTF) forms the core of the NMTF-DTI approach.
• The method incorporates multiple kernels for drug and target similarity, alongside Laplacian regularization to improve predictive performance.

Main Results:

• NMTF-DTI demonstrated superior performance compared to existing DTI prediction methods across four gold standard datasets.
• Evaluation metrics, including Area Under the ROC Curve (AUC) and Area Under the Precision-Recall Curve (AUPR), confirmed the method's effectiveness.
• A case study further validated the superiority of NMTF-DTI in predicting drug-target interactions.

Conclusions:

• NMTF-DTI is a highly effective computational tool for predicting drug-target interactions.
• The proposed method offers a significant advancement in accelerating drug discovery and development pipelines.
• NMTF-DTI's superior performance suggests its potential to reduce the time and cost associated with identifying potential drug candidates.