Skip to main content
. 2020 Feb 11;9:e51271. doi: 10.7554/eLife.51271

Figure 1. High-throughput single-cell transcriptomic profiling of the habenula.

(A) Schematic for scRNA-seq using the inDrop platform. Tissue containing the habenula was microdissected from acute coronal brain slices prepared from adult mice (1). Tissue chunks were digested in a cocktail of proteases and followed by trituration and filtration to obtain a cell suspension (2). Single cells were encapsulated using a droplet-based microfluidic device (3) for cell barcoding and mRNA capture (4). RNA sequencing (5) and bioinformatics analysis followed (6). (B) t-SNE plot of the processed dataset containing 7,506 cells from six animals. Cells are color-coded according to the cluster labels shown in (C). (C) Left: Dendrogram with cell class labels corresponding to clusters shown in (B). Right: Dot plot displaying expression of example enriched genes used to identify each major cell class. The color of each dot (blue to red) indicates the relative log-scaled expression of each gene, whereas the dot size indicates the fraction of cells expressing the gene.

Figure 1.

Figure 1—figure supplement 1. Comparison of cell-type composition across hemispheres and gene diversity, mitochondrial genes, and UMIs across cell types.

Figure 1—figure supplement 1.

(A) t-SNE plot of the dataset with cells color-coded by the hemisphere from which the sample was acquired. (B) Bar plots showing the percentage of cells in each hemisphere that are categorized into each of the 12 major cell types. (C) Violin plots of the number of genes (top), unique molecular identifiers (UMIs, middle), and percentage of mitochondrial genes per cell (bottom) for each of the 12 cell types. Each point represents a single cell and filled area is a probability distribution of all the cells in that category.
Figure 1—figure supplement 2. Expression of genes known to be important for habenular microglial and astrocytic function.

Figure 1—figure supplement 2.

Violin plots of gene expression (scaled and log normalized) for Tnf (A), Tnfrsf1a (TNF-receptor) (B), and Kcnj10 (Kir4.1) (C). Each point represents a single cell and the filled area is a probability distribution of all the cells in that category.