iCircDA-LTR

iCircDA-LTR applies a Learning to Rank (LTR) framework to predict and rank associations between circular RNAs (circRNAs) and diseases to support identification of disease-linked biomarkers.

Key Features:

• Ranking Framework: Uses a Learning to Rank (LTR) algorithm to model and rank global associations between query circRNAs and diseases, capturing ranking information among associations.
• Supervised Learning: Employs supervised learning to integrate multiple predictors and features for ranking circRNA–disease association scores.
• Performance: Validated on two independent test datasets and reported to outperform competing methods, particularly for predicting diseases associated with newly discovered circRNAs.

Scientific Applications:

• Association Prediction: Prioritizes candidate disease associations for known and newly discovered circRNAs.
• Biomarker and Target Discovery: Supports identification and prioritization of circRNA biomarkers and potential therapeutic targets in disease studies.
• Genomics Research: Provides ranked association outputs useful for genomics and bioinformatics investigations of circRNA roles in disease mechanisms.

Methodology:

Integrates various predictors and features within a supervised Learning to Rank (LTR) framework to compute ranked circRNA–disease association lists.