DTI-MLCD: predicting drug-target interactions using multi-label learning with community detection method.

Area of Science:

• Computational chemistry
• Bioinformatics
• Machine learning

Background:

• Identifying drug-target interactions (DTIs) is crucial for drug discovery and repositioning.
• Computational methods, particularly machine learning, are used to predict DTIs, but performance can be improved.
• Multi-label learning offers potential to enhance predictive accuracy by considering label correlations.

Purpose of the Study:

• To develop an advanced computational method for predicting drug-target interactions.
• To enhance the performance of drug-target interaction prediction using multi-label classification.
• To introduce a novel approach, DTI-MLCD, incorporating community detection for DTI prediction.

Main Methods:

• Applied multi-label classification with community detection algorithms for DTI prediction (DTI-MLCD).
• Utilized an updated gold standard dataset with 15,000 additional positive DTI samples.
• Compared DTI-MLCD against existing machine learning and DTI prediction methods.

Main Results:

• The DTI-MLCD method demonstrated superior performance compared to other machine learning and existing DTI prediction approaches.
• The updated dataset provided a more comprehensive basis for evaluating DTI prediction models.
• The proposed method effectively handles the complexities of multi-label learning in DTI prediction.

Conclusions:

• DTI-MLCD offers a significant advancement in computational drug-target interaction prediction.
• The integration of community detection in multi-label learning is effective for DTI prediction.
• The study provides valuable resources, including datasets and source code, for further research.