miRLAB

miRLAB infers and validates microRNA (miRNA)-mRNA regulatory relationships by applying benchmark computational prediction methods and supporting validation with experimentally validated target and perturbation data.

Key Features:

• Automated Data Collection: Automates acquisition of matched miRNA and mRNA expression profiles from The Cancer Genome Atlas (TCGA).
• Comprehensive Computational Methods: Implements 12 benchmark computational methods that use miRNA/mRNA expression data and sequence-based information to predict miRNA-mRNA regulatory relationships.
• Validation Pipelines: Validates predicted targets using experimentally validated miRNA target data and miRNA perturbation data.
• Comparative Analysis Tools: Provides functions to compare and evaluate performance across the included computational methods.
• Enrichment Analyses: Performs enrichment analyses to explore biological pathways and functional associations of predicted targets.

Scientific Applications:

• Cancer research: Analysis of miRNA-mRNA interactions in cancer datasets, including TCGA, to study gene regulation involved in tumorigenesis.
• Model benchmarking and experimental design: Benchmarking computational prediction methods and informing experimental design through validated predictions and comparative evaluations.

Methodology:

miRLAB collects matched miRNA and mRNA expression data from TCGA and applies selected computational methods (the included 12 benchmark algorithms using expression and sequence-based information) to predict regulatory relationships. Predictions are validated against experimentally validated miRNA target data, miRNA perturbation data, third-party databases, and literature.