iLncDA-LTR

iLncDA-LTR predicts associations between newly detected long non-coding RNAs (lncRNAs) and diseases to prioritize candidate disease links for biological and biomedical research.

Key Features:

• Learning to Rank (LTR) Framework: Employs a Learning to Rank-based approach that frames lncRNA–disease prediction as an information retrieval ranking problem.
• Integration of Multiple Relevant Information: Integrates various relevant pieces of information into the LTR model to enhance prediction accuracy.
• Focus on Newly Detected lncRNAs: Targets newly detected lncRNAs, including those identified by high-throughput sequencing technologies, rather than relying solely on known associations.
• Performance Evaluation: Demonstrated superior performance compared with existing state-of-the-art predictors on independent datasets.

Scientific Applications:

• Identification of disease mechanisms: Facilitates identification of potential disease mechanisms involving novel lncRNAs.
• Therapeutic target and diagnostic marker discovery: Supports exploration of new therapeutic targets and diagnostic markers linked to lncRNAs.
• Personalized medicine research: Contributes to prioritizing lncRNA–disease associations for downstream experimental validation in precision medicine contexts.

Methodology:

Treats newly detected lncRNAs as queries and diseases as documents in an information retrieval context and predicts candidate diseases by ranking candidates using an integrated Learning to Rank model.