liminfoCytoscape Reference
Free reference guide: Cytoscape Reference
About Cytoscape Reference
The Cytoscape Reference provides a comprehensive, searchable guide to the Cytoscape network analysis and visualization platform, covering network import formats (SIF, XGMML, GraphML, CSV/TSV), STRING protein interaction database integration, visual style mapping for nodes and edges, and publication-quality export options.
This reference includes advanced analysis features such as NetworkAnalyzer topology metrics (degree distribution, betweenness/closeness centrality), essential plugins (clusterMaker2 for MCL clustering, BiNGO for GO enrichment, CytoHubba for hub gene discovery, MCODE for dense module detection, Enrichment Map for pathway visualization), and layout algorithms (force-directed, hierarchical, yFiles).
The guide also covers Cytoscape.js, the JavaScript library for web-based network visualization, including initialization, data-driven styling with mapData(), event handling, and layout options (CoSE, dagre, circle, grid). Ideal for bioinformatics researchers, systems biologists, and web developers building interactive network applications.
Key Features
- Network import guide for SIF, XGMML, GML, GraphML, CSV/TSV, and STRING database queries
- Node and edge visual style mapping with continuous, discrete, and passthrough mapping types
- NetworkAnalyzer topology metrics: degree, betweenness centrality, closeness centrality, clustering coefficient
- Plugin references for clusterMaker2 (MCL), BiNGO (GO analysis), CytoHubba (MCC ranking), and MCODE
- Force-directed, circular, hierarchical, and yFiles layout algorithms with parameter tuning
- Cytoscape.js API reference with styling, events, layouts, and image export for web applications
- Publication-quality image export in PNG (300 DPI), SVG vector, and PDF formats
- Enrichment Map plugin for gene set enrichment visualization with AutoAnnotate cluster labeling
Frequently Asked Questions
What network file formats does Cytoscape support?
Cytoscape supports SIF (Simple Interaction Format, the simplest tabular format), XGMML (XML-based with node/edge attributes), GML, GraphML, CX/CX2, Cytoscape.js JSON, PSI-MI, and SBML. You can also import from CSV/TSV spreadsheets by mapping source, target, and interaction columns. Additionally, the stringApp plugin enables direct import from the STRING protein interaction database with configurable confidence cutoffs.
How do I create data-driven visual style mappings in Cytoscape?
In the Style panel, select a visual property (e.g., Fill Color, Size, Shape) and choose a mapping type. Continuous mapping maps numeric data (like log2FC or degree) to a color gradient or size range. Discrete mapping assigns categorical values (like node_type) to specific shapes or colors. For example, map log2FC from blue (negative) through white (zero) to red (positive) for gene expression heatmap-style coloring.
What does NetworkAnalyzer compute and how do I interpret the results?
NetworkAnalyzer (Tools > Analyze Network) computes degree (connection count), betweenness centrality (frequency on shortest paths, identifying bottleneck nodes), closeness centrality (inverse average distance to all nodes), clustering coefficient (neighbor connectivity), network diameter, and connected components. Results are added as node/edge table columns and can be mapped to visual styles for intuitive visualization.
How do I find hub genes using CytoHubba?
Install CytoHubba from the App Manager, then run Apps > CytoHubba. Select a ranking method such as MCC (Maximal Clique Centrality, considered most accurate), Degree, Betweenness, or Closeness. Choose the top N nodes (e.g., 10). The plugin colors hub nodes from red (highest) to yellow and extracts them as a subnetwork, useful for drug target candidate and biomarker discovery.
What is the difference between force-directed and yFiles layouts?
Prefuse Force Directed Layout is the most commonly used physics-based layout where nodes repel and edges act as springs. yFiles layouts are commercial-quality algorithms: Organic Layout minimizes node overlap for large networks, Hierarchic Layout clearly shows signaling pathways with top-to-bottom arrangement, Orthogonal Layout creates circuit-diagram style with right-angle edges, and Radial Layout arranges nodes concentrically around a hub.
How do I perform Gene Ontology enrichment analysis in Cytoscape?
Install BiNGO from the App Manager. Select your nodes of interest, then run Apps > BiNGO. Choose the organism, ontology (Biological Process recommended), statistical test (Hypergeometric), and multiple testing correction (Benjamini-Hochberg FDR) with a p-value cutoff of 0.05. BiNGO generates a GO hierarchy network where significant terms are colored from yellow (slightly significant) to orange (highly significant).
How do I use Cytoscape.js for web-based network visualization?
Initialize Cytoscape.js by creating a cytoscape instance with a container element, elements array (nodes and edges with data objects), style array (selectors with CSS-like properties), and a layout name. Use mapData() for data-driven styling (e.g., mapping score to color gradient). Handle events with cy.on("tap", "node", callback). For layouts, CoSE is recommended for general use, while dagre (requires extension) provides hierarchical arrangement.
How do I export high-quality network images for publications?
Use File > Export > Network to Image. For publications, choose SVG format for vector graphics that maintain quality at any scale, or PNG at 300 DPI resolution with transparent background option. In Cytoscape.js, use cy.png({scale: 2, full: true}), cy.svg({scale: 1, full: true}), or cy.jpg({quality: 1}). Sessions can be saved as .cys files preserving all network data, styles, and layout positions.