Modeling the functional network of primary intercellular Ca2+ wave propagation in astrocytes and its application to study drug effects.

We introduce a simple procedure of multivariate signal analysis to uncover the functional connectivity among cells composing a living tissue and describe how to apply it for extracting insight on the effect of drugs in the tissue. The procedure is based on the covariance matrix of time resolved activity signals. By determining the time-lag that maximizes covariance, one derives the weight of the corresponding connection between cells. Introducing simple constraints, it is possible to conclude whether pairs of cells are functionally connected and in which direction. After testing the method against synthetic data we apply it to study intercellular propagation of Ca(2+) waves in astrocytes following an external stimulus, with the aim of uncovering the functional cellular connectivity network. Our method proves to be particularly suited for this type of networking signal propagation where signals are pulse-like and have short time-delays, and is shown to be superior to standard methods, namely a multivariate Granger algorithm. Finally, based on the statistical analysis of the connection weight distribution, we propose simple measures for assessing the impact of drugs on the functional connectivity between cells.