Computational prediction and analysis of histone H3k27me1-associated miRNAs

Area of Science:

• Epigenetics and Molecular Biology
• Computational Biology
• Genomics

Background:

• Histone modifications, specifically mono-methylation of histone H3 on lysine 27 (H3K27me1), are crucial for cellular processes.
• H3K27me1 directly influences microRNA (miRNA) transcription by interacting with miRNA DNA sequences, highlighting its regulatory role.

Purpose of the Study:

• To develop and validate a machine learning model for predicting H3K27me1-associated miRNAs.
• To investigate the biological functions and potential clinical relevance of H3K27me1-regulated miRNAs through enrichment analyses.

Main Methods:

• A machine learning approach was employed to predict miRNAs associated with H3K27me1.
• Leave-one-out and five-fold cross-validation were used to assess model performance, achieving AUCs of 0.6866 and 0.6849, respectively.
• Enrichment analyses were conducted on transcription factors, Gene Ontology (GO) terms, and pathways associated with the predicted H3K27me1-miRNAs.

Main Results:

• The computational method demonstrated moderate predictive accuracy for H3K27me1-associated miRNAs.
• Enrichment analysis revealed that several transcription factors linked to H3K27me1-miRNAs are unfavorable prognostic markers in renal cancer.
• Associated miRNAs were found to be involved in RNA and protein binding, crucial for transcriptional and translational regulation, and implicated in cancer, signaling, and viral pathways.

Conclusions:

• The study presents a novel computational tool for identifying H3K27me1-regulated miRNAs.
• H3K27me1-associated miRNAs play significant roles in gene regulation and are implicated in various cellular processes, including cancer development and progression.