MiRComb

MiRComb integrates miRNA and mRNA expression with hybridization information to identify context-specific miRNA–mRNA interactions in disease, reflecting miRNA binding to 3' untranslated regions (UTRs) that can lead to mRNA degradation.

Key Features:

• R package implementation: Implemented as an R package for computational analysis of miRNA–mRNA interactions.
• Integration of expression data and hybridization information: Combines miRNA and mRNA expression profiles with hybridization sequences to refine prediction of miRNA targets.
• Context-specific filtering: Applies filters that consider physiological or pathological conditions to prioritize interactions relevant to the studied context and reduce false positives.
• Filtering of prediction databases: Filters candidate interactions coming from existing miRNA target prediction databases to prioritize biologically supported pairs.
• Standardized output: Produces results in a standardized PDF report format.
• Demonstrated on TCGA gastrointestinal cancers: Applied to The Cancer Genome Atlas (TCGA) data for five gastrointestinal cancers: colon cancer, rectal cancer, liver cancer, stomach cancer, and esophageal cancer.

Scientific Applications:

• Oncology research: Analyze miRNA–mRNA interactions to study regulatory mechanisms specific to different cancer types.
• Hypothesis generation: Prioritize candidate miRNA–mRNA pairs to generate testable hypotheses for experimental validation.
• Integrative cancer-type analyses: Conduct comparative analyses across multiple cancer types to identify shared or unique regulatory patterns.
• miRNA regulatory network investigation: Investigate miRNA-mediated regulatory networks by linking expression data with target hybridization evidence.

Methodology:

Combines miRNA and mRNA expression profiles with hybridization information, applies contextual filters based on physiological or pathological conditions, and filters results against existing miRNA target prediction databases.