Predicting the functions of long noncoding RNAs using RNA-seq based on Bayesian network.

Area of Science:

• Genomics
• Molecular Biology
• Bioinformatics

Background:

• Long noncoding RNAs (lncRNAs) are crucial in biological processes, but their functions remain largely unknown.
• Accurate functional annotation of lncRNAs is essential for understanding gene regulation and disease mechanisms.

Purpose of the Study:

• To develop a computational framework for predicting lncRNA functions.
• To construct a regulatory network integrating lncRNAs and protein-coding genes.
• To identify novel biological roles of lncRNAs using RNA-sequencing data.

Main Methods:

• Transcript profiles of lncRNAs and protein-coding genes were generated using RNA-sequencing data.
• A Bayesian network approach was employed to build a regulatory network, inferring dependency relationships.
• Functional enrichment analysis, combined with protein-protein interaction networks, was used to predict lncRNA functions.

Main Results:

• The framework successfully assigned functions to 762 lncRNAs in prostate cancer RNA-sequencing data.
• Predicted lncRNA functions include involvement in diverse biological processes such as tissue and embryonic development (e.g., nervous system and mesoderm development).
• The method demonstrated comparable accuracy to existing approaches like neighboring gene-based inference and lncRNA knockdown experiments.

Conclusions:

• The developed framework provides a robust method for predicting lncRNA functions from RNA-sequencing data.
• This approach facilitates the identification of complex lncRNA-gene interactions and uncovers critical lncRNA roles in biological pathways.
• The method is applicable to new RNA-sequencing datasets, aiding researchers in functional genomics studies.