Fig. 1

Hierarchical clustering of single-cell expression profiles identifies ten subgroups in the head. a Overview of the single-cell RNAseq pipeline. Planarian heads were amputated at the posterior extent of the brain lobes and dissociated. Cells were stained with Hoechst and single cells were isolated by FACS. cDNA libraries were prepared and sequenced and reads were aligned to the SmedASXL transcriptome assembly. Cluster and pseudotime analyses were subsequently performed. b Transcripts were plotted according to their mean expression level and coefficient of variation across all single-cell samples and ranked by significance of their deviation from the fit. The top 1000 highly variable transcripts are circled in red. The solid line is the regression line; dashed lines indicate the 95 % confidence interval. c Hierarchical clustering (HC) based on the top 1000 highly variable transcripts identified ten subgroups in the planarian head (colored boxes, group number indicated below). d, e t-Distributed Stochastic Neighbor Embedding (t-sne) plots of single cells colored by HC group membership (d) or FACS gate (e). f Heatmap of log2 normalized gene expression for the top 1000 highly variable transcripts in each single cell sample. Color bars: HC group, group membership colored as in (d); FACS gate, colored as in (e); alpha-tubulin, yellow indicates detection; piwi-1 and piwi-2 normalized expression counts are in greyscale according to the color key at the upper left