Dual balanced augmented topological noncoding RNA disease triplet association in heterogeneous graphs

Area of Science:

• Genomics
• Bioinformatics
• Computational Biology

Background:

• Noncoding RNAs (ncRNAs), including long noncoding RNAs (lncRNAs) and microRNAs (miRNAs), are critical in human diseases.
• Predicting associations like lncRNA-disease (LDAs) and miRNA-disease (MDAs) is vital for understanding disease mechanisms and therapeutic targeting.
• Existing models struggle with data imbalance and lack specificity in predicting diverse ncRNA-disease and ncRNA-ncRNA interactions.

Purpose of the Study:

• To propose the Dual Balanced Augmented Topological Noncoding RNA Disease triplet Association (DBATNDA) model.
• To overcome limitations of existing methods in handling data imbalance and providing differentiated predictions for specific ncRNA types.
• To enable fast, accurate, and specific predictions of ncRNA-disease and ncRNA-ncRNA triplet associations.

Main Methods:

• Constructed an Interaction Dual Graph incorporating LDAs, MDAs, and lncRNA-miRNA interactions (LMIs).
• Implemented a graph-based balanced topological augmentation mechanism to improve node representation and handle imbalanced data.
• Utilized a node classification approach for predicting triplet associations.

Main Results:

• DBATNDA demonstrated superior performance compared to state-of-the-art models in predicting ncRNA-disease and ncRNA-ncRNA triplet associations.
• The model effectively addresses data imbalance issues inherent in biological association prediction.
• Case studies confirmed the practical significance and target specificity of DBATNDA's predictions.

Conclusions:

• DBATNDA offers a novel dual-representation strategy for simultaneous, differentiated prediction of diverse ncRNA-disease and ncRNA-ncRNA associations.
• The model enhances accuracy and specificity in predicting these crucial biological interactions.
• DBATNDA represents a significant advancement in computational approaches for noncoding RNA research and disease association studies.