GCNPCA: miRNA-Disease Associations Prediction Algorithm Based on Graph Convolutional Neural Networks.

Area of Science:

• Bioinformatics
• Computational Biology
• Genomics

Background:

• MicroRNAs (miRNAs) play a crucial role in human diseases, with aberrant expression linked to disease onset and progression.
• Identifying disease-associated miRNAs as biomarkers advances disease pathology and clinical medicine.
• Experimental validation of miRNA-disease correlations is costly and time-consuming, necessitating efficient computational prediction methods.

Purpose of the Study:

• To develop an accurate and efficient computational algorithm for predicting miRNA-disease associations.
• To leverage deep learning and dimensionality reduction techniques for enhanced prediction accuracy.

Main Methods:

• A novel algorithm, Graph Convolutional neural Networks and Principal Component Analysis (GCNPCA), was developed.
• Graph Convolutional Neural Networks (GCN) with an attention mechanism extracted topological information from a heterogeneous miRNA-disease network.
• Principal Component Analysis (PCA) processed internal node attributes, and Random Forest (RF) classified combined features.

Main Results:

• The GCNPCA algorithm demonstrated high predictive performance.
• Five-fold cross-validation yielded an Area Under the Curve (AUC) of 0.983 and an Area Under the Precision-Recall Curve (AUPR) of 0.988.

Conclusions:

• GCNPCA offers an efficient and accurate computational approach for predicting miRNA-disease associations.
• The method effectively integrates topological and attribute information for robust biomarker discovery.