The term network or system has become one of the most popular concepts in a number of sciences, from computer engineering to biology. Although in all these disciplines a network is defined by different properties and is subject to a different set of assumptions, the term generally implies a common structure which graphically represents a mechanism described by one or more functions.
In a biological sense, a network refers to a set of reactions expressing a particular process in a biological system by means of genomic particles, motifs, modules, and their interactions. Hereby in order to better understand the genomic data, we need to detect the underlying components by means of computational tools such as by z-score and p-value approaches for the motif detection or by modularity coefficients and shortest path length methods for the module investigation.
On the other side, it is known that the biological networks can also show a great variety in terms of the type of connections and structures of nodes such as their scale-freeness, randomness, or modularity. Thereby to separate different networks and to compare them, several quantitative criteria are suggested. We call these criteria as the topological features which consist of the (i) degree distribution, (ii) clustering coefficient, (iii) characteristic path length and diameter, (iv) existence of hubs and network robustness, (v) flux of reactions, and (vi) existence of hierarchical modularity.
Due to the distinction of these features in the evaluation of different networks, their detections’ algorithms and testing procedures for their statistical significances can be helpful to describe the general pattern of genomic connectivities in a system and to critically assess the statements about the network topology.