Bipartite graph-based collaborative matrix factorization method for predicting miRNA-disease associations.

Area of Science:

• Biotechnology
• Genomics
• Computational Biology

Background:

• MicroRNAs (miRNAs) are critical in human diseases.
• Experimental identification of miRNA-disease associations (MDAs) is costly and time-consuming.
• Computational methods are increasingly used to predict novel MDAs.

Purpose of the Study:

• To propose an efficient computational method for predicting novel miRNA-disease associations (MDAs).
• To leverage bipartite graphs and collaborative matrix factorization for improved MDA prediction.

Main Methods:

• Developed a novel method named bipartite graph-based collaborative matrix factorization (BGCMF).
• Employed collaborative matrix factorization on a bipartite graph structure.
• Utilized improved recommendation techniques within the model.

Main Results:

• The BGCMF method achieved a high Area Under the Curve (AUC) of 0.9514 ± 0.0007 in five-fold cross-validation.
• The model demonstrated superior performance compared to existing state-of-the-art methods.
• Successfully predicted numerous potential associations between novel miRNAs and diseases.

Conclusions:

• The BGCMF method is effective for predicting potential miRNA-disease associations.
• The high AUC value validates the method's predictive capability.
• BGCMF offers a valuable tool for advancing research in human diseases by identifying new miRNA roles.