Statistical analysis of synchrony and synchrony propagation in massively parallel spike trains

The mechanisms by which the brain represents, processes and transmits information interms of spiking activity are largely unknown. The coordination of spikes among neuronsat millisecond temporal scale has been hypothesized as a mechanism of information coding.Spike synchronization is considered a signature of such cooperative dynamics. Experimentalstudies have found evidence of the occurrence of spike synchronization beyond chancelevel in relation to behavior. However, these findings were mostly limited to data froma small number of simultaneously recorded neurons, and were concluded on the basis ofpairwise or low-order correlation analyses unable to conclude on the presence of neuronalinteractions of any order.Recent advances in electrophysiological technology enable recordings of hundreds ofneurons in parallel, a number which is destined to grow. The increase in data dimensionalitymakes traditional methods for spike correlation analysis inapplicable due to thecombinatorical explosion in the number of patterns, which raises severe computational andstatistical problems. New tools for the analysis of big data are thus needed.This essay delivers two novel methods for the analysis of spike synchronization in massivelyparallel spike trains. The first method, called SPADE (spike pattern detection andevaluation) is designed to perform an optimal search for synchronous spike patterns usingefficient data mining techniques. The detected patterns are then tested for significanceusing a hierarchy of statistical tests that circumvents multiple testing issues.We employ SPADE to investigate the occurrence of spike synchronization in the motorcortex of two macaque monkeys performing an instructed-delay reach-to-grasp task. Thetask consists of different epochs and was performed in different ways. By analyzing eachepoch and trial type separately, we are able to associate the emergence of individual patternsof synchronous spikes to specific behaviors, and we investigate the spatial organizationof synchronous spiking neurons in motor cortex. Our analysis reveals groups of neurons, aligned along a preferential medio-lateral orientation, which selectively synchronize theiractivity in relation to behavior.We finally introduce a second statistical method, called ASSET (analysis of sequencesof synchronous events), that searches for temporal sequences of synchronous spike patterns. Synchrony propagation has been long suggested as a plausible mechanism of informationcoding and transmission, successfully implemented for instance in the synfire chain modelin a manner compatible with electrophysiological constraints. Here we provide, to thebest of our knowledge, the first statistical tool to detect sequential synchronous events inmassively parallel spike trains.In summary, this work delivers two novel powerful statistical tools, SPADE and ASSET,to analyze spike synchrony and synchrony propagation in parallel spike trains. Thesemethods are designed for application to large data from hundreds or thousands of simultaneousrecordings, and cope with typical features of real data, such as non-stationarity offiring rates in time and across neurons or regularity in the spiking activity. The applicationof SPADE to motor cortex data from two macaque monkeys performing a reach-to-grasptask allows us to shed new light on the spatial and temporal organization of spike synchronizationin the cerebral cortex in relation to behavior