Cell type prioritization in single-cell data

We present a machine-learning method to prioritize the cell types most responsive to biological perturbations within high-dimensional single-cell data. We validate our method, Augur (https://github.com/neurorestore/Augur), on a compendium of single-cell RNA-seq, chromatin accessibility, and imaging transcriptomics datasets. We apply Augur to expose the neural circuits that enable walking after paralysis in response to spinal cord neurostimulation. We devised an experiment to expose the neuron subtypes recruited by TESS using single-cell transcriptomics. Mice received a severe contusion of the thoracic spinal cord that led to permanent paralysis of both legs. In the presence of 5-HT1a and D1 agonists, TESS immediately enabled walking in paralyzed mice. We performed single-nucleus RNA-seq of 20,484 nuclei from mice placed on the treadmill for 30 min with or without TESS, identifying all the major cell types of the lumbar spinal cord. We then subjected the 6,171 identified neurons to an additional round of clustering, revealing the presence of 38 neuron subtypes that expressed classical marker genes and were detected across experimental conditions.