iRefScape

iRefScape integrates and visualizes consolidated protein interaction data within Cytoscape to facilitate analysis of redundancy, overlapping evidence, high interconnectivity, and disease-related gene networks.

Key Features:

• iRefIndex consolidation: Leverages the iRefIndex consolidation of protein interaction information from ten distinct databases using sequence-based hash keys to reduce redundancy and merge overlapping records.
• Uniform accession space: Employs a uniform accession space that minimizes data redundancy and supports search expansion across multiple accession types.
• Data filtering and mining: Provides node and edge attributes to enable programmatic and analytical filtering and mining of interaction networks.
• Provenance information: Encodes provenance using node colors and additional attributes to trace original source evidence.
• Multi-graph representation: Uses a multi-graph approach to represent overlapping evidence and multiple interaction datasets simultaneously.
• Bi-partite representation: Distinguishes binary and n-ary source data through bi-partite graph representations.
• Disease-related gene searches (OMIM): Supports targeted searches and analysis of disease-related genes cataloged in OMIM.
• Adjacency-matrix view: Provides a synchronized adjacency-matrix view for visualizing relationships between user-defined groups of proteins or biological entities.

Scientific Applications:

• Interaction network analysis: Enables analysis of protein–protein interaction networks to address redundancy, overlap, and high interconnectivity in consolidated datasets.
• Exploratory knowledge discovery: Supports searches ranging from targeted retrieval to exploratory discovery across merged interaction sources.
• Disease gene network interrogation: Facilitates investigation of OMIM-listed disease-related genes within consolidated interaction networks.
• Source-provenance comparisons: Allows tracing and comparison of original source evidence across multiple interaction databases via provenance annotations.

Methodology:

Consolidation via iRefIndex from ten databases using sequence-based hash keys; use of a uniform accession space with search expansion across accession types; multi-graph and bi-partite graph representations to manage overlapping evidence and binary versus n-ary data; node and edge attributes and node-color provenance annotations for filtering/mining; synchronized adjacency-matrix for group relationship visualization.