A sub-optimal criterion to estimate and compare neural spiking activity on Micro-Electrode Array technology

In neurobiology an even bigger role is covered by acquisition and processing of data obtained from micro-electrode array (MEA) technology, where a fixed number of channels record in vitro neural spiking activity. In this work we developed a software-tool implementing a specific method to analyze time data series of this spiking activity. Our aim is to quantify a global measure of correlation and dependence between activities of neurons belonging to different MEA channels. Methods include cross-correlation, mutual information and dynamic time warping. These techniques can extract specific information about behavior of the neurons matrices. In order to synthesize a global measure of correlation in spiking neuron activity, we applied a sub-optimal criterion based on genetic algorithms (GA): GA help us to sort MEA channels with most similar and coordinated activity, and finally to estimate our index of global similarity. Thanks to this index we tried to test the existence of a sort of self-coordinating neuronal network, evolving in time and regulating the ensemble neural behavior under specific external stimuli, such as the chemical ones here studied