Spatiotemporal patterns of neuronal activity are hypothesized to be essential for information processing in the brain. They suggest the existence of cell assemblies, groups of co-active neurons that represent a distinct cognitive unit, potentially related to specific behaviors or representations. Conventionally, cell assemblies are defined by strong structural connections, such that the stimulation of a portion of its members should transiently activate the entire assembly. Their exact composition and computational role, however, have yet to be fully clarified. We present the detection of spike patterns across superficial and deep layers of rat motor cortex in a voluntary forelimb-movement task, extending on a previous report of diverse activation of pyramidal neurons over different sequential motor phases. In our approach, snapshots of neuronal activity are compared with flexible temporal alignment, relying on an extension of the "edit similarity score", a metric originally introduced to compare strings. We further investigated the participation of neurons in these flexible patterns, hereby named "profiles", through graph analysis. Across animals, profiles spanned both layers and occurred preferentially, but not exclusively, close to moments of reward. By connecting neurons in weighted graphs of co-participation in profiles, we observed non-trivial structures with effective hubs that were not explained by shuffled models. While detected profiles were not representative of entire experimental sessions, specific nodes (neurons) and edges (neuron pairs) were persistent over time. We argue that beyond synchronous activation, neurons are organized in a temporal sequence space, in which profiles with strong overlap group them together. Individual profiles belonging to a community in this space can therefore be understood as different realizations of an underlying functional group, an extension of the concept of cell assembly with structural and temporal flexibility.