MiRNA-disease association prediction via hypergraph learning based on high-dimensionality features.

Area of Science:

• Biomedical Informatics
• Computational Biology
• Genomics

Background:

• MicroRNAs (miRNAs) play a crucial role in various human complex diseases.
• Identifying disease-related miRNAs is vital for understanding disease pathogenesis.
• Experimental methods for miRNA-disease association identification are costly and time-consuming.

Purpose of the Study:

• To develop an efficient computational model for predicting potential miRNA-disease associations.
• To leverage high-dimensionality features and hypergraph learning for improved prediction accuracy.

Main Methods:

• The HFHLMDA method integrates miRNA functional similarity and disease semantic similarity.
• A hypergraph is constructed using the K-Nearest-Neighbor (KNN) method to represent complex miRNA-disease relationships.
• A hypergraph learning model is employed to calculate miRNA-disease association scores.

Main Results:

• HFHLMDA achieved superior performance compared to four state-of-the-art models.
• Validation demonstrated high accuracy with 92.09% (leave-one-out) and 91.87% (fivefold cross-validation).
• Case studies showed high confirmation rates for predicted associations (e.g., 90-98% for top 50 predictions).

Conclusions:

• The proposed HFHLMDA method is effective for predicting miRNA-disease associations.
• This computational approach offers a valuable tool for accelerating the discovery of disease-related miRNAs.