Abstract—The understanding of the system level of biological behavior and phenomenon variously needs some elements such as gene sequence, protein structure, gene functions and metabolic pathways. Challenging problems are representing, learning and reasoning about these biochemical reactions, gene and protein structure, genotype and relation between the phenotype, and expression system on those interactions. The goal of our work is to understand the behaviors of the interactions networks and to model their evolution in time and in space. We propose in this study an ontological meta-model for the knowledge representation of the genetic regulatory networks. Ontology in artificial intelligence means the fundamental categories and relations that provide a framework for knowledge models. Domain ontology’s are now commonly used to enable heterogeneous information resources, such as knowledge-based systems, to communicate with each other. The interest of our model is to represent the spatial, temporal and spatio-temporal knowledge. We validated our propositions in the genetic regulatory network of the Aarbidosis thaliana flower. Keywords—Ontological model, spatio-temporal modeling, Genetic Regulatory Networks (GRNs), knowledge representation. I. INTRODUCTION HE analysis of genetic regulatory networks, responsible for cell differentiation and development in prokaryotes and eukaryotes, will much benefit from the recent up scaling to the genomic level of experimental methods in molecular biology. One of the hottest research topics in Genome Science is the interaction between genes. Genetic Regulatory Network (GRN) [1] is one of the recent focuses to understand metabolic pathways and bioprocesses. GRNs act as analog biochemical computers to specify the identity and level of expression of a group of targeted genes. Its output is the constellation of RNAs (Ribonucleic acid) and proteins encoded by target genes. Time series expression data obtained from DNA microarrays is one of the most useful kinds of data used to construct and test GRNs. There are numerous techniques to model GRNs or the behavior of a cell: boolean networks, Petri nets, Bayesian Authors are with the Laboratoire de Recherche en Informatique Arabisée et Documentique Intégrée (R.I.A.D.I), Ecole Nationale des Sciences de l’Informatique. Campus Universitaire de Manouba, 2010 Manouba, Tunis, Tunisie (phone: 216 71 600 444; fax: 216 71 600 449; e-mail: ines.hamdi@riadi.rnu.tn, mohamed.benahmed@riadi.rnu.tn). networks, cluster analysis etc. There are even genetic/metabolic circuit networks where genes, metabolic enzymes, and proteins are modeled as nodes with the relationship between activation, inhibition and mediation as links. GRN models can be used to identify genetic diseases and estimate the effects of medications. Actually there few systems that are interested to the GRN modeling. Despite this systems described explicitly the modeling used techniques; they not define how represent the biological knowledge required to model the GRN. The goal of our project is to study the dynamic of the GRN by giving a spatio-temporal model, able to represent the gene expression evolution caused by external factors. Biomedical knowledge is encapsulated in tens of millions of publications with various degrees of coherence and computability. The most difficulty is to identify the biological knowledge and in particular the GRN knowledge [2]. The ontology’s are a very powerful formalisms of representation knowledge domains as complex and rich that the cellular biology. This paper will be focused to describe how resolve the knowledge representation problem in biological domain and spatially of genetic regulatory networks. We used an ontological approach by giving a GRN Ontology Design pattern (ODP). This ODP defines a spatial and temporal attributes to express the spatial and temporal knowledge’s. Our ODP is validated actually by the ARABIDOPSIS THALIANA GRN. This paper is organized as follows: section II describes the Knowledge representation in Biological Domain; section III describes the proposed approach; section IV presents our conclusions. II. THE KNOWLEDGE REPRESENTATION IN BIOLOGICAL DOMAIN Genes are complex structures and they cause dynamic transformation of one substance into another during the whole life of an individual, as well as the life of the human population over many generations. When genes are “in action”, the dynamics of the processes in which a single gene is involved are complex, as this gene interacts with many other genes, proteins, and is influenced by many environmental and developmental factors. The complexity of biological phenomena is primarily caused by interactions of biochemical components in the underlying bimolecular regulatory networks at different layers The Knowledge Representation of the Genetic Regulatory Networks Based on Ontology Ines Hamdi, and Mohamed Ben Ahmed T World Academy of Science, Engineering and Technology International Journal of Computer and Information Engineering Vol:3, No:6, 2009 1517 International Scholarly and Scientific Research & Innovation 3(6) 2009 scholar.waset.org/1307-6892/11214 International Science Index, Computer and Information Engineering Vol:3, No:6, 2009 waset.org/Publication/11214