MLWNNR: LncRNA-Disease Association Prediction with Multi-Kernel Learning-Driven Weighted Nuclear Norm Regularization.

Area of Science:

• Genomics
• Bioinformatics
• Computational Biology

Background:

• Long non-coding RNAs (lncRNAs) are key regulators in human diseases.
• Predicting lncRNA-disease associations is crucial for understanding disease mechanisms.

Purpose of the Study:

• To develop a robust algorithm for forecasting lncRNA-disease associations.
• To leverage multi-kernel learning and network completion for accurate predictions.

Main Methods:

• Utilized a k-nearest neighbors-based kernel learning algorithm to integrate multi-similarity kernels.
• Constructed a heterogeneous lncRNA-disease association network.
• Applied weighted nuclear norm regularization for network completion and association scoring.

Main Results:

• The MLWNNR algorithm demonstrated superior performance on three datasets compared to six other models.
• Case studies confirmed a majority of predicted lncRNA-disease associations with existing literature.
• The model exhibited robustness and excellent generalization capabilities.

Conclusions:

• MLWNNR provides a reliable computational approach for inferring lncRNA-disease associations.
• This method can aid in identifying novel biomarkers and therapeutic targets for human diseases.