Figure 4
From: Integrating single-cell RNA-seq and imaging with SCOPE-seq2
Application of SCOPE-seq2 to a human GBM surgical sample. (A) UMAP embedding of the cell scores from scHPF factorization of the scRNA-seq data colored based on unsupervised clustering from Phenograph. (B) Same as (A) but colored by scHPF cell scores for each scHPF factor. A short list of top-scoring genes for each factor is also included. (C) Identification of imaging meta-features. A heatmap shows the z-scored values of 16 cell imaging features (columns) across cells (rows), and a dendrogram indicates three feature clusters, cell size, shape and Calcein staining intensity, from an unsupervised hierarchical clustering. (D) Heterogeneity of cell imaging meta-features. Boxplots show the distribution of imaging meta-features in each Phenograph cluster from scRNA-seq.
