GNAT (Glycosylation Network Analysis Toolbox)

Introduction

      GNAT is a free open-source MATLAB toolbox. It is:

  • Written using MATLAB and Java.      
  • Supported in Windows, Linux and Mac OS platforms.
  • Available from http://sourceforge.net/projects/gnatmatlab/files/.
  • Individual GNAT commands can be implemented from the MATLAB command prompt. 
  • Custom MATLAB .m scripts can be written to integrate GNAT functions and classes into other code.Toolbox Demos and the GettingStarted.pdf file enclosed with the toolbox provide sample scripts.



Feature 
GNAT contains a variety of classes to describe glycans and glycosylation reaction networks. It also provides various methods to manipulate these classes.The most recent release of  GNAT (GNATv2beta.zip) provides a variety of functions that can be categorized in seven groups as below: 

Network Input/Output
These functions enable integration of XML (eXtensible Markup Language) based glycan structures into SBML (Systems Biology Markup Language) format files. These functions also allow reading & writing of individual glycan and reaction network files.

Display Glycans, Enzymes and  Networks 
functions enable visualization of individual glycans and glycosylation reaction networks in the MATLAB environment. Updated enzyme classes are now able to handle biochemical data on enzyme specificity including functional group,linkage and substrate specificity. These enzyme classes can also be displayed using GUI.
 
Database Query
These functions enable linkage between GNAT and primarily two databases called GlycomeDB and the Consortium of Functional Glycomics (CFG).


Network Inference
These functions enable single-reaction inference and also automated full network reconstruction. The input for such algorithms include either sets of reactants and/or products along with the list of enzymes present in a system.
 
Graph Operations
These functions enable a variety of network manipulation methods. These include algorithms to find the path between any two glycans in the network,  to generate the subset network model and to examine the network connectivity. 

Mass Spectrometry Analysis
Several functions are provided in order to enable the synthesis of reaction networks from mass spectrometry data.

Simulation
These functions enable dynamic and steady-state simulation of reaction networks.

Installation
A brief instruction for the installation of GNAT is provided as follows.  Detailed step-by-step instructions are provided in the GettingStarted.pdf document enclosed in the package.  

  • Start MATLAB. MATLAB version 7.13 (R2011b) or higher is recommended. 
  • Install libSBML5.6.0 or above in MATLAB environment (if not already installed). 
  • Install SBML toolbox 4.1.0 or above (if not already installed). 
  • Install GNAT in MATLAB. To do this, download the GNAT package and extract files into the installation directory (say: C:\gnat). In the MATLAB command windows, go to this installation directory, and type "installGNAT". This will install the software along with all necessary paths. 
  • Change java class path in MATLAB. To do this, open the 'startup.m' file located in the toolbox\startup folder. Edit 'GNATInstallationDIR' (~10th line) to reflect the absolute directory path (e.g. 'C:\gnat').Copy the 'startup.m' file into the MATLAB startup folder.

Note: If the MATLAB path is set/installed properly the first time, GNAT will be fully functional. If a 'file cannot be found' error appears in subsequent MATLAB restarts,then the path set in the original installation was not saved when MATLAB was quit. One way to resolve this issue is to simply go to the GNAT installation directory and type 'installGNAT' again. For more information about other hints, please see the 'FAQ' section of the GettingStarted.pdf file.


Usage guide 
Detailed instructions for usage, examples and demos are provided in GettingStarted.pdf. 


Support
For questions about the software usage, comments or suggestions please contact Sriram Neelamegham at This email address is being protected from spambots. You need JavaScript enabled to view it.