Figure 3. SNAP-seq reveals heterogeneity of tissue adipocytes from iWAT.

(A) Workflow showing DAPI-stained nuclei pre- and post-FACS/filtration that underwent microfluidic partitioning and library preparation in the 10X genomics platform followed by sequencing using an Illumina HiSeq 4000. (B) tSNE-plot showing 14 clusters from ~6000 adipocytes derived from iWAT of mice exposed to cold for 24 hr. Each colored dot is an adipocyte assigned to a cluster based on transcriptomic signature. (C) Normalized expression values of the top two adipocyte subtype-specific cluster genes from (B) plotted as violin plots with clusters as rows and genes as columns. (D) tSNE-plot showing cluster-specific expression of selected marker genes from (C). (E) Normalized expression values of indicated genes in subtype-specific clusters plotted as violin plots with clusters as rows and genes as columns. Black arrow is pointing toward metabolically active Type nine adipocyte cluster and enriched gene.

Figure 3—figure supplement 1. Multimodal analysis of SVF and adipocyte single-cell sequencing data.

Hierarchical plot showing correlation of single cell/nuclei gene expression between SVF and adipocyte clusters. Cell types from both mature adipocyte nuclei and SVF single cells were used in this analysis. They were grouped using hierarchical clustering with tiles colored by unadjusted (top) -Log10 Bonferroni adjusted p-values (bottom). Unadjusted (top) and adjusted (bottom) p-values were thresholded to aid in visualization with values less than 10⁻⁵ set to 10⁻⁵.