RNMFLP: Predicting circRNA-disease associations based on robust nonnegative matrix factorization and label...

Area of Science:

• Genomics
• Bioinformatics
• Computational Biology

Background:

• Circular RNAs (circRNAs) are stable noncoding RNA molecules with significant roles in human diseases.
• Experimental validation of circRNA-disease associations is resource-intensive and time-consuming.
• There is a need for efficient computational methods to predict these associations.

Purpose of the Study:

• To develop and validate a novel computational method, RNMFLP, for predicting circRNA-disease associations.
• To improve the accuracy and efficiency of identifying disease-related circRNAs.
• To provide a reliable tool for circRNA-disease association discovery.

Main Methods:

• The RNMFLP method integrates robust nonnegative matrix factorization (RNMF) and label propagation (LP) algorithms.
• It refines the circRNA-disease adjacency matrix using integrated circRNA and disease similarity.
• RNMF captures latent features, while LP predicts associations from similarity networks.

Main Results:

• RNMFLP demonstrated superior performance compared to existing state-of-the-art methods across four datasets via fivefold cross-validation.
• The method accurately predicted circRNA-disease associations, as confirmed by case studies.
• Case studies on lung cancer, hepatocellular carcinoma, and colorectal cancer validated the method's reliability.

Conclusions:

• RNMFLP is a powerful and reliable computational tool for predicting circRNA-disease associations.
• The method offers a cost-effective and efficient alternative to experimental validation.
• RNMFLP facilitates the discovery of novel disease-related circRNAs, advancing our understanding of disease mechanisms.