Temporal computation by synaptic signaling pathways

Abstract

Synaptic signaling comprises a complex molecular network. Such networks carry out diverse operations such as molecular logic, signal amplification, memory and other aspects of cellular decision-making ([Bray, 1995]). The synapse in particular encounters complex input patterns that have different temporal sequences. Different input patterns to the synapse are known to give rise to a range of synaptic responses, including facilitation, depression and various forms of short and long-term potentiation. In many cases the stimuli that generate these disparate responses are tens of seconds or more in length, much greater than the typical time-courses of calcium dynamics. In this paper I propose that the synaptic signaling network can perform temporal computation operations such as tuning for stimulus duration or interval. Using simulation methods I show that the simple time-courses of individual signaling pathways combine in the network to give rise to different temporally selective responses. Downstream pathways that exhibit temporal integration or amplitude thresholding select different input patterns and thus perform temporal computation.