CatbNet
Tools for comparing and analyzing the topology of biological networks
Getting Started
CatbNet is able to compute most important network topological features, from node-based measures such as: different centralities to whole network-based measures such as: network density, diameter, for analytical and comparative purposes. You can use this tool by downloading its python source code. This tool is created using python 2.7
Prerequisites
If you are interested in using python source code, consider installing packages:
Networkx version 1.9.1
Numpy
Matplotlib
Pandas
Scipy
PyQt4
Running and Loading Data
To run CatbNet from python source run command:
python main.py
In CatbNet there is the opportunity to laod data from .gml, .graphml, .net and .el network data formats. In the case of loading .el (edge list) networks you should consider that:
- Each line must demonstrate a node or an edge
- Data must be tab seperated
- In weighted networks, each edge line must have three values in a line
- Nodes and Edges are seprated by #nodes and #edges keywords
example:
#nodes
a
b
c
#edges
a b 1.45
a c 2.6
Considerations
In the case you want to compare network groups, you should check option 'network files are classified in groups'.
Test Application
As a Test Data, you can test network files of grouped networks in directory: test-data
License
This project is licensed under the LBB License - It is free to use for only researching goals. For more details contact:
Ehsan Pournoor
Email: e.pournoor@ut.ac.ir
Laboratory of Bioinformatics and Systems Biology (LBB)