Kirillova O.V., Samsonova M.G., Serov V.N.
Institute of High Performance Computing and Data Bases, 198005, St.Petersburg, Fontanka 120, office 7, Russia; Fax +7-812-251-8314, E-mail samson@fn.csa.ru
Understanding the mechanisms of cell functioning requires the investigation of complex behaviour of ensembles of interacting genes. Traditional biochemical methods are highly biased towards the identification of single genes and their effects. Thus the elucidation of cell function should call for novel approaches which go beyond the traditional experimental methods. One of this approaches is to consider the ensembles of interacting genes as a genetic networks. The simplest and the most computational effective model system that gives insights into the overall behaviour of large genetic networks is the Boolean network. In this model genes are represented as the elements of the Boolean net, and the wiring of the elements to one another correspond to the functional links between genes. The visualisation and user interaction are the important tools and techniques which assist scientists in evaluation and absorbtion of data. As the genetic networks consist of tens and perhaps hundreds of genes involved in complex regulatory interactions it is necessary to develop special purpose computational tools for their visualization. Moreover deciphering of structure, logic and complex behavior of large genetic networks will require the development of tools for simulation of their dynamics. We design the Java applet NetWork (http://www.csa.ru/Inst/gorb_dep/inbios/Dyn_bool/Dyn.htm) which enables the user to construct interactively the genetic network of interest, as well as to evaluate its dynamics in framework of Boolean network model. The user is presented with the display of genetic network in the form of oriented graph: genes are represented as rectangles, and their interactions - as arrows. For the presentation of genes and their interactions we modified the NODES and EDGES classes developed by Sun Microsystems Inc. A new class DYNAMICS was developed which contains the method for simulation of network dynamics based on the Boolean algebra technique. There are only two parameters in the HTML file containing the applet. One of them points on the location of the help file, while the other defines the name of the main applet class. All other aspects of data presentation and applet functionality are controlled by the user, who inputs the genetic network, optimizes its visualization and simulates the genetic network dynamics. NetWork displays the mechanism of genes interactions in the network and enables visualization of large genetic networks. The applet enables the user to identify rapidly and to explore attractors. Using NetWork it is possible to model the effects of the mutations in the network, to reveal gene interactions compensating for these mutations, as well as to adjust the genetic network structure to fit biological data.