An integrative network-driven pipeline for systematic identification of lncRNA-associated regulatory network motifs in metastatic melanoma

摘要

An integrative network-driven pipeline for discriminating non-metastatic and metastatic melanoma phenotypes based on lncRNA-associated regulatory network motifs. a Intermolecular interaction data between regulatory molecules (lncRNA, miRNA, and gene/TF) are extracted from public databases, literature, and predicted using existing tools (RNAhybrid, RPISeq). b Interactions are merged together to generate an integrated network. Topological properties of the network are investigated with the Network Analyzer plugin in Cytoscape. Non-topological properties, including disease pathway association retrieved from KEGG database and expression profiles of the nodes obtained from a melanoma-specific patient dataset from UCSC Xena. c Important regulatory loops comprising of lncRNA-miRNA-TF are predicted with the help of NetDS Cytoscape plugin. d Network motifs are prioritized using a multi-objective function by providing user-defined weights in an iterative manner. e Calculation of motifs prediction accuracy and survival analysis