IHNLncSim

IHNLncSim infers functional similarity among long non-coding RNAs (lncRNAs) by integrating miRNA interaction data, disease associations, GTEx expression profiles, and NONCODE expression data into a combined network for lncRNA function prediction.

Key Features:

• miRNA-based Similarity Network: Utilizes miRNA interaction data to assess lncRNA functional similarity.
• Disease-based Similarity Network: Considers associations between lncRNAs and diseases to define similarity.
• GTEx Expression-based Network: Incorporates GTEx gene expression profiles as a basis for similarity.
• NONCODE Expression-based Network: Uses expression data from the NONCODE database to derive similarity.
• Integrated Network Construction: Combines four single lncRNA functional similarity networks into a novel integrated network.
• Performance Evaluation: Reports cross-validation Area Under the Curve (AUC) of 0.736 for the integrated network versus miRNA-based 0.703, disease-based 0.733, GTEx expression-based 0.611, and NONCODE expression-based 0.602.
• Disease-based Enrichment Analysis: Provides disease-based lncRNA function enrichment analysis as an analytical module.
• Shared Target mRNA Inference: Infers functional similarities based on shared target mRNAs.

Scientific Applications:

• Large-scale lncRNA Function Prediction: Enables large-scale prediction of unknown lncRNA functions using integrated similarity metrics.
• Regulatory Mechanism Investigation: Supports investigation of lncRNA regulatory mechanisms by identifying functionally similar lncRNAs through shared target mRNAs and disease associations.

Methodology:

Constructs an integrated network from four single lncRNA functional similarity networks (miRNA interactions, disease associations, GTEx expression, NONCODE expression) and evaluates model performance by cross-validation reporting AUC metrics.