Multiscale characteristics of cell proliferation in the developing central nervous system of chick embryos

MAZZEO, JORGE; RAPACIOLI, MELINA; RODRÍGUEZ CELÍN ALEJANDRA; DUARTE, SANTIAGO; FLORES, VLADIMIR

Punta del Este, Uruguay

Congreso; XVI Conference on Nonequilibrium Statistical Mechanics and Nonlinear Physics (XVI Medyfinol 2008); 2008

The spatial organization of cell proliferation (CP) is studied within the mathematical framework ofstochastic point processes. Numerical series were built recording the Euclidean distances betweensuccessive proliferating cells along the cephalic-caudal axis of the optic tectum of chick embryos.The main goal of this work is to assess interactions between proliferating cells. The centralhypothesis is that, if proliferating cells behave interactively, such interactions should install somekind of dependency or memory on the signals representing the spatial organization of theproliferative activity. This dependency may be revealed by the value of a scaling index. Severalalgorithms were applied for computing the estimates: Rescaled Range Analysis, DetrendedFluctuation Analysis, Fano Factor, Power Spectral Density, Dispersional Analysis. Althoughnumerical series representative of CP are non-stationary due to the spatial asymmetric distribution of the proliferation process, the values of scaling indexes estimated by several methods correspond to signals possessing stationary correlations. It is known that the non-stationarity may cause this discrepancy; i.e., several authors highlighted that scaling estimator algorithms may give biased results on non-stationary series. Global trends removal is a usual approach to overcome this pitfall. Analyses of detrended CP series allow the detection of anticorrelation immersed in CP original series. The anticorrelation may be interpreted as the existence of local short range inhibiting interactions between proliferating neuroepithelial cells.