Abstract
In plants, chromatin accessibility – the primary mark of regulatory DNA – is relatively static across tissues and conditions. This scarcity of accessible sites that are dynamic or tissue-specific may be due in part to tissue heterogeneity in previous bulk studies. To assess the effects of tissue heterogeneity, we apply single-cell ATAC-seq to A. thaliana roots and identify thousands of differentially accessible sites, sufficient to resolve all major cell types of the root. However, even this vast increase relative to bulk studies in the number of dynamic sites does not resolve the poor correlation at individual loci between accessibility and expression. Instead, we find that the entirety of a cell’s regulatory landscape and its transcriptome each capture cell type identity independently. We leverage this shared information on cell identity to integrate accessibility and transcriptome data in order to characterize developmental progression, endoreduplication and cell division in the root. We further use the combined data to characterize cell type-specific motif enrichments of large transcription factor families and to link the expression of individual family members to changing accessibility at specific loci, taking the first steps toward resolving the direct and indirect effects that shape gene expression. Our approach provides an analytical framework to infer the gene regulatory networks that execute plant development.
Competing Interest Statement
The authors have declared no competing interest.
Footnotes
Major changes: - New analyses that directly compare scATAC-seq epidermal cell data with INTACT ATAC-seq of root hair and root non-hair cell types have been added. - New endoreduplication analysis has been added to account statistically for total UMI count differences across cells. - A new Supplementary Figure 2 has been added to show the patterns of traditional marker genes using both scATAC and scRNA data sets; Supplemental Table 3 has been added to show marker genes for each cell type. - Major revisions to the Methods section have been added to explain the developmental progression and endoreduplication metrics described in the manuscript. Updated Figure panels: To better reflect the structure of the manuscript, Supplementary Figures have been reordered as follows: Original Figure S1 -> revised Figure S1 Original Figure S2 -> revised Figure S3 Original Figure S3 -> revised Figure S5 Original Figure S4 -> revised Figure S6 Original Figure S5 -> revised Figure S4 Original Figure S6 -> revised Figure S7