SIMNETS: a computationally efficient and scalable framework for identifying networks of functionally similar neurons

Abstract

The expansion of large-scale neural recording capabilities has provided new opportunities to examine multi-scale cortical network activity at single neuron resolution. At the same time, the growing scale and complexity of these datasets introduce new conceptual and technical challenges beyond what can be addressed using traditional analysis techniques. Here, we present SIMNETS, a mathematically rigorous and efficient unsupervised relational analysis framework designed to generate intuitive, low-dimensional neuron maps that support a multi-scale view of the computational similarity (CS) relations among individual neurons. The critical innovation is the use of a novel measure of computational similarity that is based on comparing the intrinsic structure of latent spaces representing the spiking output of individual neurons. We use three publicly available neural population test datasets from the visual, motor, and hippocampal CA1 brain regions to validate the SIMNETS framework and demonstrate how it can be used to identify putative subnetworks (i.e., clusters of neurons with similar computational properties). Our analysis pipeline includes a novel statistical test designed to evaluate the likelihood of detecting spurious neuron clusters to validate network structure results. The SIMNETS framework can facilitate linking computational geometry representations across scales, from single neurons to subnetworks, within large-scale neural recording data.