KATZNCP: a miRNA-disease association prediction model integrating KATZ algorithm and network consistency projection

Area of Science:

• Bioinformatics
• Computational Biology
• Genomics

Background:

• MicroRNAs (miRNAs) play a crucial role in human health and disease development.
• Understanding miRNA-disease associations is vital for disease prevention and treatment.
• Computational prediction of miRNA-disease links complements experimental validation.

Purpose of the Study:

• To propose a novel computational model, KATZNCP, for predicting potential miRNA-disease associations.
• To leverage the KATZ algorithm and network consistency projection for enhanced prediction accuracy.
• To provide a valuable tool for guiding future experimental research in miRNA-disease interactions.

Main Methods:

• Constructed a heterogeneous network integrating known miRNA-disease associations, miRNA similarities, and disease similarities.
• Applied the KATZ algorithm to the heterogeneous network to generate initial miRNA-disease prediction scores.
• Utilized network consistency projection to refine scores and obtain final predictions.

Main Results:

• The KATZNCP model achieved a high Area Under the Curve (AUC) of 0.9325 in leave-one-out cross-validation (LOOCV).
• Performance of KATZNCP surpassed existing state-of-the-art comparable algorithms.
• Case studies on lung and esophageal neoplasms confirmed the model's excellent predictive capability.

Conclusions:

• KATZNCP is an effective computational model for predicting potential miRNA-disease associations.
• The model integrates network analysis with established algorithms for robust predictions.
• KATZNCP offers valuable guidance for experimental validation and therapeutic development.