Heterogeneous network model to infer human disease-long intergenic non-coding RNA associations

Area of Science:

• Molecular Biology
• Genomics
• Computational Biology

Background:

• Long non-coding RNAs (lncRNAs) are crucial regulators in biological processes like cell differentiation and disease.
• Understanding disease-lncRNA associations is vital for biological research and human health.

Purpose of the Study:

• To develop a novel computational method, KRWRH, for inferring disease-long non-coding RNA associations.
• To leverage phenotype information and lncRNA tissue expression data for improved association prediction.

Main Methods:

• Utilized Gaussian interaction profile kernel for diseases and lncRNAs.
• Employed a random walk with restart algorithm for final prediction.
• Incorporated phenotype information and tissue expression details of lncRNAs.

Main Results:

• KRWRH demonstrated superior performance in predicting disease-lncRNA associations compared to existing methods.
• Validation through leave-one-out cross-validation, ROC curves, and mean enrichment confirmed KRWRH's effectiveness.
• The method successfully identified both known and previously unknown disease-lncRNA relationships.

Conclusions:

• KRWRH offers a powerful and effective computational approach for disease-lncRNA association inference.
• The integration of phenotype and tissue expression data enhances prediction accuracy.
• This method holds potential for advancing our understanding of lncRNA roles in human diseases.