RepTar

RepTar: miRNA Target Prediction in 3'-UTRs

RepTar identifies repetitive elements within 3'-untranslated regions (3'-UTRs) of genes to predict miRNA–mRNA interactions independent of evolutionary conservation and canonical Watson-Crick seed pairing.

Key Features:

• Modular Algorithm: Detects repetitive elements in 3'-UTRs without requiring evolutionary conservation or conventional Watson-Crick pairing of seed regions.
• Diverse Binding Site Prediction: Predicts conventional seed sites, non-conventional sites with seed wobbles (G-U pairing within the seed region), 3'-compensatory sites, and centered sites.
• Conservation-Independent Targeting: Identifies cellular targets of viral miRNAs without reliance on evolutionary conservation.
• Genome-wide Predictions: Provides genome-wide target predictions for human and mouse miRNAs, including cellular targets of human and mouse viral miRNAs.

Scientific Applications:

• miRNA Regulatory Analysis: Characterizes conventional and non-conventional miRNA binding sites to investigate miRNA-mediated gene regulation and viral miRNA targeting mechanisms.

Methodology:

Applies a modular computational framework to scan 3'-UTRs for repetitive sequence elements and classify predicted miRNA binding sites into conventional seed, seed wobble (G-U pairing), 3'-compensatory, and centered site categories without incorporating evolutionary conservation constraints.