RNA-Seq of single cells from the Arabidopsis root

The definition of cell identity is a central problem in biology. Single-cell RNA-seq provides a wealth of information regarding the developmental state of individual cells. However, better methods are needed to map the identity of single cells, especially during identity transitions. We have developed a quantitative classification method that is robust to expression noise and can detect primary and chimeric identities from single-cell RNA sequencing profiles. The method uses existing transcriptome repositories of grouped cell-types to define a set of optimal cell-identity markers, which are then used to define a cell identity metric. This metric accurately classified diverse cell identities in Arabidopsis root tips and human glioblastoma cells. We demonstrate the strength of the approach to resolve a dynamic developmental process by analyzing the identity of single cells captured from regenerating Arabidopsis roots following removal of their stem-cell-niche. We discover that, apart from new niche formation at the vicinity of the cut site, cells that are distant from the injury site also undergo a transient, partial collapse of identity during the regeneration and reorganization of the root, demonstrating the usefulness of a quantitative cell identity metric.