Analysis of node centrality measures in complex biological networks

Document Type : Original Article

Authors
Ali Sarabadani 1
M. Lapechi Dizji
K. Rahseparfard 2
Mehrdad Saffarie 3
1 Qom University
2 Qom University, Qom, Iran
3 Qom university
Abstract
The interpretation of the results of biological networks should always depend on the topological study of the nodes, currently there is no consensus on how to use these criteria, and most network analyzes always lead to a basic interpretation of a limited number of criteria. A biological network is any network that is used for biological systems. A network is a system that includes sub-parts that are connected to each other in the majority of a whole, such as different species that are related to each other in the food web of the whole. To fully understand biological networks, a coherent understanding of the concept of node centrality is essential. Therefore, for 10 typical node metrics in biological networks, this study first evaluates their current applications, advantages, disadvantages, as well as their potential applications. Then, an overview of previous studies is provided and correspondingly, suggestions are provided to improve biological topology algorithms. Finally, the following recommendations are presented in this study:
(1) A comprehensive and accurate evaluation of node centrality requires the use of multiple criteria, including both the target node and its surrounding neighbors, and it can be calculated from the maximum neighborhood density component, as used in addition to other criteria of node centrality; (2) different centrality measures can be used to identify nodes with different functions, which are depicted in this study as modular adjacencies, bridging roles, and aptitude. and (3) the following groups of node centrality can often be validated against each other, including degree and maximum neighborhood component, oddness, closeness and radius, stress, and betweenness.