TDAview

TDAview implements topological data analysis using the Mapper algorithm to construct and visualize Mapper graphs from high-dimensional biological datasets for applications such as topology-based gene co-expression network analysis.

Key Features:

• Mapper algorithm: Implementation of the Mapper algorithm to construct graphs representing the shape of high-dimensional data.
• Parameterizable graph construction: Supports specification of parameters to tailor the construction of Mapper graphs.
• Dual-mode functionality: Analysis-and-visualization mode constructs Mapper graphs from input data and metadata while visualization-only mode imports and displays pre-generated Mapper graphs.
• Metadata visualization: Graphical display of metadata variable values associated with nodes and edges of Mapper graphs.
• Publication-quality output: Exports high-quality figures suitable for publication.
• Scalability: Handles large datasets with tens of thousands of data points.
• Topology-based gene co-expression networks: Enables construction and analysis of topology-based gene co-expression networks.

Scientific Applications:

• Genomics: Construction and exploration of topology-based gene co-expression networks to investigate gene co-expression patterns, genetic interactions, and regulatory mechanisms.
• High-dimensional biological data analysis: Topological summarization and visualization of complex biological datasets including multi-variable metadata.
• Cross-domain complex data analysis: Exploration of complex high-dimensional datasets in domains beyond biology using Mapper-based topology.

Methodology:

Implements the Mapper algorithm to construct parameterizable Mapper graphs from input data and metadata, supports importing pre-generated Mapper graphs, displays metadata values on nodes and edges, and is optimized to handle datasets with tens of thousands of points.