CompositeView: A Network-Based Visualization Tool

Area of Science:

• Bioinformatics
• Network Science
• Data Visualization

Background:

• Complex network analysis is crucial in fields like bioinformatics and knowledge graphs.
• Existing tools often struggle with information overload and lack dynamic score calculation.
• Interactive visualization aids in extracting actionable insights from large datasets.

Purpose of the Study:

• Introduce CompositeView, a Python-based open-source application for enhanced complex network visualization.
• Develop a tool that simplifies information overload through auto-calculated composite scores.
• Enable interactive data refinement for better insight extraction.

Main Methods:

• CompositeView utilizes specifically formatted input data to calculate composite scores.
• It employs the Cytoscape component of Dash for visualization.
• Interactive filtering (node value, edge weight) and graph manipulation refine results.

Main Results:

• CompositeView effectively reduces information overload by generating composite scores.
• The application demonstrates versatility with network and non-network data, including biomedical literature, HDI, and cardiovascular study data.
• Performance benchmarks were established for data visualization breadth and size.

Conclusions:

• CompositeView offers a novel approach to interactive network visualization and data analysis.
• Its auto-calculation and updating of composite scores differentiate it from existing tools.
• The application is a valuable tool for fields requiring complex network analysis and insight extraction.