Fig. 4

Human β-cell subtypes using unsupervised scRNA-seq and snRNA-seq data integration. A Integrated and self-annotated clusters on UMAPs of scRNA-seq and snRNA-seq data. B Canonical endocrine cell gene marker expression on scRNA-seq and snRNA-seq integrated dataset. C Separated β-cell cluster from integrated main data identifying three cell sub-clusters using a Louvain resolution of 0.4. D INS expression pattern in the β-cell cluster in scRNA-seq and snRNA-seq data on UMAP. E INS expression level on violin plots in the three β-cell subtypes in the scRNA-seq and snRNA-seq data. F Monocle 3 generated pseudo-time dimensional reduction plot in the separated β-cell cluster from integrated main data using lower INS expression area in scRNA-seq data (cluster 3) as base. G Re-annotated β-cell sub-clusters of scRNA-seq and snRNA-seq data based on pseudo-time INS expression. H ZNF385D expression pattern in the β-cell sub-clusters in scRNA-seq and snRNA-seq data on UMAP. I ZNF385D expression level on violin plots in the three β-cell subtypes in the scRNA-seq and snRNA-seq data. J HNRNPA2B1 expression pattern in the β-cell sub-clusters in scRNA-seq and snRNA-seq data on UMAP. K HNRNPA2B1 expression level on violin plots in the three β-cell subtypes in the scRNA-seq and snRNA-seq data