“…Examples of these approaches include computational strategies to capture "free" information from existing data, including developmental trajectories (e.g., Monocle 71 , Palantir 72 , SlingShot 73 , and CellRank 74 ), RNA dynamics (e.g., RNA velocity 75 and scVelo 76 ), methods to integrate disparate and multimodal datasets (e.g., Seurat 77 , Harmony 78 , Symphony 79 ), methods that implement differential gene expression analysis (e.g., MILO 80 ), and several end-to-end pipelines that implement large collections of tools in a single computational ecosystem (e.g., Seurat 81 , Monocle 71 , scanpy 82 ). While many of the tools already developed for analysis of single-cell data from mammalian tissues will be applicable to analysis of plant data sets, and indeed some tools were specifically developed for the analysis of plant singlecell data (e.g., Asc-Seurat 43 , COPILOT 20 , Socrates 33 ), several computational challenges remain that are unique or of particular consideration to plants and would benefit from the development of still more new tools and databases. For example, as research moves from Arabidopsis roots in the first wave of studies to new species and tissues, how do we know if cell clusters represent true cell types if no high-quality cell type-specific markers are already known?…”