Protein-driven inference of miRNA-disease associations

Area of Science:

• Biochemistry
• Genetics
• Bioinformatics

Background:

• MicroRNAs (miRNAs) are crucial non-coding RNAs regulating cellular functions and implicated in various diseases.
• Current knowledge on miRNA-disease associations is limited in scope and reliability.
• Existing tools lack the ability to propose molecular mechanisms underlying these associations, hindering experimental validation.

Purpose of the Study:

• To develop a computational method, miRPD, for inferring direct microRNA-protein-disease associations.
• To facilitate hypothesis generation regarding the molecular basis of miRNA-related diseases.
• To provide a reliable resource for researchers investigating miRNA functions in disease.

Main Methods:

• Coupling known and predicted miRNA-protein interactions with text-mined protein-disease associations.
• Developing scoring schemes to rank inferred miRNA-disease associations by confidence levels (high and medium).
• Utilizing literature mining to extract protein-disease relationships.

Main Results:

• The miRPD database explicitly infers miRNA-protein-disease associations, suggesting underlying proteins for hypothesis generation.
• Scoring schemes enable the creation of reliable high- and medium-confidence miRNA-disease association sets.
• Analysis revealed significant enrichment of proteins in cancer and type I diabetes mellitus pathways, indicating potential biological trends or literature bias.

Conclusions:

• miRPD provides a valuable resource for exploring miRNA-disease links and their molecular underpinnings.
• The inferred associations can guide experimental validation and deepen understanding of disease pathogenesis.
• The findings suggest specific pathways potentially influenced by miRNAs in cancer and diabetes.