This topic is also a subject of this paper. Kim et al. in his paper developed a concept of temporally varying networks. Each time-specific network has its own network motifs and the network motifs change over time. Temporal change of the network structure means that a static network, i.e., the network derived from binding experiments, representing logical relationships between genes, is utilized differently at different times during some time- evolving process. If we imagine the dynamic nature of gene expression, where expression of particular genes changes over time, then the different temporal patterns of the networks shown in Figure 1 represent temporal gene expression levels in the form of Cyclosporine a network diagram. In principle, Figure 1 can be redrawn to a movie with the snapshots shown in Figure 2. In Figure 2, the shading of a gene node and its connection reflects the influence of the regulator on the temporal expression level of the regulated gene. The concept of varying networks is thus a projection of gene expression dynamics in the form of a directed graph of gene interactions. By examining the temporary gene expression profiles, it is obvious that at a particular moment, the expression of a particular gene can be so low that the connection to this node is practically functionless. Evolution from one state of the potential network to another over time is graphically depicted in Figure 2. It is obvious from these analyses that the networks derived from static DNA binding experiments are only potential and that their temporal realization depends on the state of gene expression at a given time point. Genetic networks can, in principle, be described by a directed graph. Such modeling invokes a Boolean relationships among the nodes of a network; that is, if gene A is connected with gene B by a logical relationship, then if A is ON, Clofentezine B is also ON or OFF. For these networks, it is quite easy to calculate terminal states as attractors or basins of attraction, and from this point of view, they have been extensively studied. In the real world, the situation is more complicated because gene expression is, in principle, a set of binding equilibria and biochemical reactions; thus, the expression level of a regulated gene depends on the expression level of the regulator. This notion led to the introduction of logical and threshold functions to the Boolean networks, which made Boolean networks more realistic, but it was more difficult to determine the parameter values of a given function. In addition to the Boolean approaches, transcrip- tional networks have been modeled using a variety of other methods, such as Bayesian networks, Petri nets or, recently, Gaussian processes. Genetic network models are summarized in several reviews. Genetic networks represent causal relationships among regula- tors and regulated genes, which can also be regulators. Such interaction then form complex networks with feedback and feed forward loops whose topology have been quite extensively studied in recent years. To what extent the dynamics of gene expression can influence the network properties is the subject of this paper. If we want to formalize transcription control processes so that they can be treated mathematically, then we can start with fundamental molecular interactions that lead to gene transcrip- tion. In principal, the probability of occurrence of a gene transcription event is given by the probability of binding of a given transcription factor molecule to the promoter region of a gene. Other molecules can be considered as readily available in sufficient amount, and therefore, referring to the principles of chemical reaction kinetics, the determining factor in the process of transcription is the number of molecules of a particular transcription factor that is present.