CLING

CLING prioritizes candidate cancer-related long non-coding RNAs (lncRNAs) by integrating multiple lncRNA-centric biological networks to identify likely cancer-associated lncRNAs.

Key Features:

• Integration of Multiple Biological Networks: Integrates multiple lncRNA-centric networks and leverages the topological properties of lncRNAs to capture complex interactions.
• Joint Optimization and Prioritization: Performs joint optimization across integrated networks to prioritize all candidate lncRNAs for each cancer type.
• Validation and Performance: Validation studies reported Area Under Curve (AUC) scores between 0.85 and 0.94 across ten cancer types, outperforming single-network prioritization methods.
• Discovery of Novel lncRNAs: Identified novel candidate lncRNAs ranked among top candidates that were subsequently confirmed by biological experiments.
• Complementarity to Differential Expression Analyses: Detected novel cancer-related lncRNAs in liver hepatocellular carcinoma that were missed by differential expression analyses (DEA).

Scientific Applications:

• Candidate Prioritization: Narrows thousands of lncRNA candidates to prioritized lists for experimental validation and follow-up.
• Biomarker and Therapeutic Target Discovery: Ranks likely cancer-associated lncRNAs to support identification of biomarkers and therapeutic targets.
• Integrative Functional Analysis: Enables integrative analysis of lncRNA roles in oncogenesis by combining multiple network-derived relationships.

Methodology:

CLING integrates multiple lncRNA-centric networks, analyzes lncRNA topological properties, and applies joint optimization to produce cancer-specific lncRNA rankings, with performance evaluated by AUC metrics (0.85–0.94) across ten cancer types.