LncRNA-miRNA interaction prediction through sequence-derived linear neighborhood propagation method with information...

Area of Science:

• Bioinformatics
• Molecular Biology
• Genomics

Background:

• Long non-coding RNAs (lncRNAs) regulate microRNAs (miRNAs), impacting complex diseases.
• Identifying lncRNA-miRNA interactions is vital for understanding lncRNA function.
• Existing computational methods for predicting these interactions are limited.

Purpose of the Study:

• To propose a novel computational method for predicting lncRNA-miRNA interactions.
• To enhance the accuracy and efficiency of identifying these molecular interactions.
• To provide a tool for discovering novel lncRNA-miRNA relationships.

Main Methods:

• Developed a sequence-derived linear neighborhood propagation method (SLNPM).
• Calculated integrated lncRNA-lncRNA and miRNA-miRNA similarities using known interactions and sequence data.
• Constructed similarity-based graphs and employed label propagation for scoring potential interactions.
• Proposed two SLNPM editions: SLNPM-SC and SLNPM-PC, based on different information combination strategies.

Main Results:

• SLNPM-SC and SLNPM-PC demonstrated higher accuracy in predicting lncRNA-miRNA interactions compared to state-of-the-art methods.
• The methods successfully identified novel lncRNA-miRNA interactions in case studies.
• Both known interactions and sequence information proved valuable for prediction.

Conclusions:

• The proposed SLNPM methods (SLNPM-SC and SLNPM-PC) are effective for predicting lncRNA-miRNA interactions.
• Known interaction data is the most critical factor, with sequence data offering supplementary information.
• These computational tools show promise for advancing research in lncRNA and miRNA biology.