Introduction

NETSCOPE is a MATLAB/Octave/Python toolbox for information theoretical analysis of molecular networks. It can be used for network construction and analysis from a wide variety of biological data. Applications range from constructing gene co-expression networks and using topological patterns to identify genes or pathways of interest, to identifying functional links in fMRI- and EEG-based networks. The toolbox is plug-and-play, just download the code and start importing your data.

Download and Setup

Download repository with:

git clone https://github.com/DepartmentofNeurophysiology/NETSCOPE.git

In MATLAB/Octave: to add the toolbox files to the search path and enable all functionality, navigate to the NETSCOPE directory and run the startup command.

Documentation

Type help <function> to see detailed information about a function.

Supporting data

Network construction from synthetic data (Figure 3 code)

synthetic_data.m contains everything necessary to recreate Figure 3 from the manuscript. The code generates synthetic expression data from a ground truth network, constructs an MI-based network and compares it to the ground truth network. The results are shown as figures.

Network construction from yeast expression data (Figure 5-7 data)

yeast_data.mat contains the gene expression data from Ziemann et al., and the co-expression network data from YeastNet v3. See Table 1 from the manuscript for more details on the network data.

Manuscript

Bergmans T, Jamal T, Rezeika A, Hsing C, Celikel T. 2026. “NETSCOPE: Information-Theory Based Network Discovery and Analysis” (in submission)