Predicting novel CircRNA-disease associations based on random walk and logistic regression model

Area of Science:

• Genomics
• Bioinformatics
• Computational Biology

Background:

• Circular RNAs (circRNAs) are noncoding RNA molecules that regulate gene expression.
• Dysregulated circRNAs are implicated in various diseases, serving as potential biomarkers.
• Experimental identification of circRNA-disease associations is costly and time-consuming.

Purpose of the Study:

• To develop an efficient computational model for predicting novel circRNA-disease associations.
• To overcome the limitations of experimental methods for circRNA-disease association discovery.

Main Methods:

• Constructed a circRNA-circRNA similarity network using Gene Ontology.
• Applied random walk with restart for feature extraction.
• Utilized a logistic regression model (RWLR) for prediction.

Main Results:

• RWLR achieved higher AUC values in Leave-One-Out Cross-Validation (LOOCV), 5-fold cross-validation (5CV), and 10-fold cross-validation (10CV) compared to PWCDA and DWNN-RLS.
• Case studies validated the reliability and effectiveness of the RWLR model.

Conclusions:

• The RWLR model offers a significant improvement in predicting circRNA-disease associations.
• RWLR provides a valuable computational tool for identifying potential disease biomarkers.