Characteristics of Complex Networks in Neural Networks

The complex network characteristics of the neuronal functional network composing of individual neurons have been rarely studied, within even fewer recording the number of neurons. In this paper, we use a multi-electrode recording spike trains datasets recorded from slice cultures of rodent somatosensory cortex in monkeys. The number of neurons in a single recording process reached several hundred. The entire recording time was divided into non-overlapping time segments to construct a functional network over different times. We analyze the small world characteristics, the modular network and community structures of neuronal functional networks. The network characteristics are analyzed to determine the changing process with the evolution of different neuronal functional networks. We proposed a new similarity coefficient S to compare the similarity indexes between the community structures. We found that neuronal networks have small world and modularity characteristics in common, when compared to a random network. The network characteristics are stronger when the networks retain a small number of edges. These findings contribute to the study of the neuronal network organization of individual neurons.