Skip to main content
Fig. 1 | Genome Biology

Fig. 1

From: Challenges and opportunities to computationally deconvolve heterogeneous tissue with varying cell sizes using single-cell RNA-sequencing datasets

Fig. 1

Diagram of example deconvolution experiment using cell scale factors. A Heatmaps of gene expression: (i) for the (y-axis) marker genes G by cell labels for each of (x-axis) neurons, oligodendrocytes, or astrocytes, (ii) the (y-axis) G marker genes by (x-axis) cell types (K). Expression value colors: blue = low, white = intermediate, red = high. (iii) Wedge diagram of (S) cell scale factors, where wedge size is the value and cartoons indicate each cell type. B (left-to-right) Heatmaps of bulk expression Y, and marker expression Z, cell scale factors S, and cell type proportions P for either (top) scaled or (bottom) unscaled expression, where bar plot values show cell type proportions with colors as in panel C. C Scatterplot of example experiment results for multiple bulk samples Y, showing the (x-axis) true cell proportions and (y-axis) predicted cell proportions, where points are outcomes for a sample and cell type, and shapes show whether the cell scale factor transformation was applied. Plots were created using the ggplot2 v3.4.1 [16] and ComplexHeatmap v2.12.1 [17] software; data used to reproduce these plots are available from GitHub (Data Availability)

Back to article page