Chapter # AN INTEGRATION OF THE DESCRIPTIONS OF GENE NETWORK

Motivation: The problems that arise when modeling complex molecular genetic systems at the cell level are so large-scale that they require integrated efforts of many research teams. Therefore, it is a topical problem to integrate the technologies applied t

86 Part 4

AN INTEGRATION OF THE DESCRIPTIONS

OF GENE NETWORKS AND THEIR MODELS

PRESENTED IN SIGMOID (CELLERATOR)

AND GENENET

Podkolodny N.L.*1, 2, Podkolodnaya N.N.1, Miginsky D.S.1, Poplavsky A.S.1,

Likhoshvai V.A.1, Compani B.3, Mjolsness E.3 1 Institute of Cytology and Genetics, SB RAS, Novosibirsk, 630090, Russia; 2 Institute of Computational Mathematics and Mathematical Geophysics, SB RAS, Novosibirsk, Russia; 3 Institute for Genomics and Bioinformatics, University of California, Irvine, USA *Corresponding author: e-mail: pnl@bionet.nsc.ru

Key words: gene network, metabolic pathway, model, databases

SUMMARY

Motivation: The problems that arise when modeling complex molecular genetic systems at the cell level are so large-scale that they require integrated efforts of many research teams. Therefore, it is a topical problem to integrate the technologies applied to description of gene networks and their models developed at the ICG (Russia) and UCI (USA) with the aim to provide a tight cooperation in the field of systems biology.

Results: A software for conversion of gene networks and the models of gene network dynamics represented in the GeneNet system into a format suitable for loading to the SIGMOID database and Cellerator system was developed.

INTRODUCTION

An original software for supporting the technological chain of modeling was designed at the Institute of Cytology and Genetics, SB RAS, including creation of databases compiling the descriptions of various organization levels of genetic systems, formalization (specification) of the models of genetic systems, study of the models’ behavior, search for the models’ parameters according to the experimentally observed gene network behavior, accumulation of basic models, and solution of the target problems. A language for the specification, SiBML, was developed; it is oriented to the construction of mathematical models of molecular genetic systems taking into account the main specific features of their structure: a linear ordering and gene orientations within the genomes, polyvariance of genes (polyallelism), and polycompartment pattern of biological systems. An original software supporting SiBML for computing the direct problem was developed as well as for solving the problem of verification of mathematical models of gene networks. An original technology GeneNet (Ananko et al., 2002, 2005) that enables accumulation of data in a database with a consequent analysis of heterogeneous information on gene and metabolic networks was designed. A large number of the gene networks describing the vital molecular genetic processes were reconstructed and are presented in the GeneNet database.

A software for systems biology SIGMOID (Cheng et al., 2005; http://doc.guandang.net/) that in turn calls Cellerator (Shapiro et al., 2003) was developed

Motivation: The problems that arise when modeling complex molecular genetic systems at the cell level are so large-scale that they require integrated efforts of many research teams. Therefore, it is a topical problem to integrate the technologies applied t

Gene networks theory: mathematical problems and software 87

at the University of California, Irvine; this software provides a wide range of options for description of biological processes and their mathematical models.

An integration of the gene networks and their models presented in GeneNet and SIGMOID will allow for an efficient combination of the technologies for modeling genetic systems developed at ICG and the technologies for distributed modeling that are developed at UCI.

THE REPRESENTATIONS OF GENE NETWORKS IN GENENET

AND SIGMOID DATABASES

Functioning of a gene network is provided by complex relationships between different components, namely, genes, proteins, metabolites, signal molecules, energy-connected cell components, etc. Using an object-oriented approach, we recognize several following logical levels in the description of relationships between the gene network components. Ontological level, including general notions and relations between them. Here we describe as metaclasses the elementary structures, or Entities (genes, proteins and protein complexes, RNAs, and small molecules) and the elementary processes (reactions and regulatory events). A scheme of semantic relationships between the elementary structures

and processes is given in Fig. 1.

Figure 1. Semantic relations between elementary notions in a gene network.

The level of objects of study involves the descriptions of notions and classes of entities for particular objects of study. For example, description of some representatives of such classes as “genes”, “proteins and protein complexes”, “RNAs”, and “small molecules”, which are involved in functioning of particular objects studied (i.e., cells of E. coli K12). The level of compartments, at which the entities described above are attached to a spatial compartment. At this level, the description of entities may include supplementary parameters such as, for example, the concentration of a given entity in a particular compartment. The whole bulk of information about elementary structures and functional relations in a gene network is represented at the three levels described above. The subsequent two levels are needed for describing the network as a whole.

The level of gene networks corresponds to the description of functional subsystems. At this level, the particular descriptions of a complex system (the object of study) from different viewpoints may be represented as well as its simplified descri …… 此处隐藏：13049字，全部文档内容请下载后查看。喜欢就下载吧 ……