Fig. 4: TAS-Seq accurately detects cell composition of murine and human lungs.
From: TAS-Seq is a robust and sensitive amplification method for bead-based scRNA-seq
a Visualization of cell clustering of each scRNA-seq dataset of the murine lung by Seurat v4.0.3 package in 2D FIt-SNE space. Note that CD4+ and CD8+ T cells were more clearly separated in FIt-SNE space in both shallow- and deep-sequenced TAS-Seq data than in the other datasets. b Stacking plot shows the composition of each annotated cell in each scRNA-seq dataset from murine lung tissue. c and d Comparison of cell composition between flow-cytometric data and scRNA-seq datasets of the murine lung. p-values of Pearson’s correlation coefficients are shown. Pearson’s correlation coefficients and slope of regression lines are shown in Supplementary Fig. 5. e Cell clustering in each scRNA-seq dataset from human RA-ILD lungs by Seurat v4.0.3 package in 2D FIt-SNE space. Note that minimal batch effects were seen between fibrotic- and non-fibrotic lung samples. f Stacking plot shows the composition of each annotated cell per scRNA-seq dataset from the human RA-ILD lung. g Comparison of cell composition between flow-cytometric data and scRNA-seq datasets of the human RA-ILD lung. p-values and R2-values of Pearson’s correlation coefficients are shown. h The composition of neutrophils of flow-cytometric and TAS-Seq data in RA-ILD lungs. i Composition of neutrophils in Smart-seq2 and 10X v2 datasets of human lungs reported by Travaglini et al. Details of the cell annotations and associated marker genes of each mouse and human dataset are shown in Supplementary Data 1.
