lncRNA-disease association prediction based on optimizing measures of multi-graph regularized matrix factorization.

Area of Science:

• Bioinformatics
• Computational Biology
• Genomics

Background:

• Long non-coding RNAs (lncRNAs) play critical roles in various biological processes and diseases.
• Accurate prediction of lncRNA-disease associations is essential for understanding disease mechanisms and developing diagnostics.
• Existing computational methods for lncRNA-disease association prediction have limitations in accuracy and scope.

Purpose of the Study:

• To develop a novel and accurate computational algorithm for predicting lncRNA-disease associations.
• To leverage multi-graph regularized matrix factorization for improved prediction performance.
• To validate the proposed algorithm against established methods using cross-validation.

Main Methods:

• Proposed a novel algorithm named optimizing measures of multi-graph regularized matrix factorization (OM-MGRMF).
• Calculated semantic similarity of diseases, functional similarity of lncRNAs, and Gaussian similarity.
• Constructed a lncRNA-disease association matrix using the K-nearest-neighbor (KNN) algorithm.
• Formulated an objective function incorporating ranking measures and multi-graph regularization constraints.
• Optimized the objective function via an adaptive gradient descent algorithm.

Main Results:

• The OM-MGRMF algorithm achieved higher prediction accuracy than classical methods.
• Experimental results demonstrated the effectiveness of the proposed approach in K-fold cross-validation.
• The method successfully integrated multiple similarity measures and regularization techniques.

Conclusions:

• The OM-MGRMF algorithm represents a significant advancement in computational prediction of lncRNA-disease associations.
• The proposed method offers a robust framework for identifying novel lncRNA-disease relationships.
• This work contributes to the field of bioinformatics by providing a more accurate tool for genomic research.