RWRMTN

RWRMTN predicts disease-associated microRNAs (miRNAs) by applying a random walk algorithm to miRNA-target gene networks to prioritize miRNAs linked to disease via mutual regulation with their target genes.

Key Features:

• Random Walk Framework: Employs a random walk algorithm on miRNA-target gene networks to capture network topology and mutual regulatory relationships between miRNAs and target genes.
• Enhanced Prediction Performance: Demonstrates improved predictive capability compared to existing network- and machine learning–based methods by exploiting mutual regulation within miRNA-target networks.
• Versatility Across Networks: Operates on any miRNA-target gene network, enabling application to diverse datasets.
• Evidence-Based Ranking: Produces ranked candidate miRNAs with support from existing literature evidence.
• Automation and Workflow Integration: Provides automation features for incorporation into external computational workflows.

Scientific Applications:

• miRNA–disease association prediction: Prioritizes miRNAs potentially associated with diseases in which miRNA dysregulation is implicated.
• Biomarker and therapeutic target identification: Supports identification of candidate miRNA biomarkers and therapeutic targets linked to disease processes.
• Oncology applications: Applied to predict miRNAs associated with breast cancer and lung cancer.

Methodology:

Constructing a network model of miRNA–target gene interactions and applying a random walk algorithm that exploits mutual regulatory relationships to predict disease associations and rank candidate miRNAs.