iSnoDi-LSGT

iSnoDi-LSGT predicts associations between small nucleolar RNAs (snoRNAs) and diseases by integrating local sequence and disease similarities with global network topological constraints.

Key Features:

• Local similarity constraints: Uses snoRNA sequence similarity and disease similarity as local constraints for association inference.
• Global topological constraints: Applies network embedding to snoRNA and disease networks to extract topological features and compute snoRNA topological similarity and disease topological similarity.
• Dual-constraint integrative model: Combines local sequence/disease similarities with global topological similarities in a unified predictive framework.
• Prediction prioritization: Produces prioritized candidate snoRNA-disease associations for downstream analysis and experimental validation.

Scientific Applications:

• Association prediction: Predicting previously unknown snoRNA-disease associations to inform studies of disease mechanisms.
• Candidate prioritization: Prioritizing snoRNAs and diseases for experimental validation and mechanistic investigation.
• Biomarker and therapeutic target discovery: Supporting identification of snoRNA-related biomarkers and potential therapeutic targets in disease contexts.

Methodology:

Combines local sequence-based similarities (snoRNA sequence similarity and disease similarity) with global network-based topological features extracted via network embedding to compute topological similarities and predict snoRNA-disease associations.