network-spatial-coherence 0.2.24

Network Validation using the Spatial Coherence Framework.

Network Spatial Coherence

What's the quality of your spatial network?

This package quantifies the spatial coherence of a network by measuring if network distances align with physical (Euclidean) distances. It helps you answer questions like: Do shortest-path network distances make spatial sense? Where, and by how much, does the network deviate from the geometric reality?

Additionally, it can reconstruct the network's original positions in space using the STRND algorithm. Networks can be simulated (if you don't have any) or imported, and weighted and bipartite networks are supported. For details, see Spatial Coherence and STRND papers.

Features

• Analyze the spatial coherence of a network
• Reconstruct images from purely network information - like drawing a country by only knowing which train stations are connected
• Efficient graph loading and processing (sparse matrices)
• Handles simulated graphs, custom graphs, unweighted graphs, weighted graphs, bipartite graphs

Install

Python 3.11 is reccomended, although older versions should work. See requirements.txt for dependencies.

pip install network_spatial_coherence

Example Results

	OG Image	SP Constant	Net Dim	Gram Mat	REC Image
Coherent Network
Incoherent Network

Quick Intro

For a quick introduction refer to Usage and Examples!

Detailed information

• Notebook
• Directory Structure
• GraphArgs Configuration
• Results Location

Interactive Network Viz

• Interactive Visualization

Citation

If you use this method or refer to its concepts in your research, please cite:

Bonet, D. F., Blumenthal, J. I., Lang, S., Dahlberg, S. K., & Hoffecker, I. T. (2024). Spatial coherence of DNA barcode networks. bioRxiv. https://doi.org/10.1101/2024.05.12.593725

Bonet, D. F., & Hoffecker, I. T. (2023). Image recovery from unknown network mechanisms for DNA sequencing-based microscopy. Nanoscale, 15(18), 8153–8157. https://doi.org/10.1039/D2NR05435C

Contact

[[e-mail]]