DiVa - Disconnectivity Valuation tool


In brief

DiVa (Disconnectivity Valuation tool) is a tool that analyzes a network by means of the pairwise disconnectivity index (PDI). The PDI estimates how significant a network element is for sustaining the pairwise connections between vertices in a network. Network elements can be vertices (genes/molecules), edges (reactions) or topological patterns (like motifs). The significance of such an element is determined by simulating its knockout from the network and comparing the number of connected ordered pairs of vertices in the network before and after the knockout. Consider the following publications for detailed information about the methodology:



For the analysis of your networks, you can either download the DiVa command line tool and make the necessary computations at your own system(s) or you can use the DiVa online webservice for a fast and automated analysis on our computer grid. The DiVa command line tool supports calculating the PDI on a single computer/CPU as well as in a computer grid to speed up the running time in case of very large networks. The DiVa online webservice is using the DiVa command line tool.


The DiVa online webservice


The DiVa online webservice requires a webbrowser with java-script enabled and is available at: http://diva.sybig.de


The DiVa command line tool



The software can be used freely except for commercial purposes. Any publication containing results that are obtained from DiVa has to acknowledge the use of the program. Please note, that there are absolutely no guarantees or warranties made by the available files for DiVa. You are working with this program at your own risk.


Installation and requirements

DiVa is implemented in C++. You can either download the pre-compiled binary or the source code to make your own one (see below). The binary has been compiled using g++ in version 4.4 on Ubuntu 12-4. Making your own binary requires a C++ compiler and the C++ boost library in version 1.46 or higher installed on your system. The boost library is available at http://www.boost.org.



Last update: October 19th, 2012.


Click here to download the binary


Click here to download the source code


Click here to download the documentation