Skip to main content
Figure 2 | BMC Bioinformatics

Figure 2

From: MCA: Multiresolution Correlation Analysis, a graphical tool for subpopulation identification in single-cell gene expression data

Figure 2

MCA plots reveal important features of the correlation structure in single-cell transcriptomics data. A. MCA plots with uniform appearance (top) reveal robust correlations amongst pairs of variables (scatter plot, bottom) like Rex1 and Sox2, sorted by Pecam1. B. Outliers can easily be detected via characteristic diagonal stripe patterns. Here a single sample with the highest value in the Sox2 distribution is enough to induce an overall positive Gbx2, Rex1 correlation (bottom, arrow). C. Robust subpopulations can be identified. The presence of a large triangular region with uniform correlation or lack of correlation between Rex1 and Nanog may indicate a subpopulation, seen here for cells from the highest 40% of the Stella distribution (top). The cells from the high Stella compartment (open boxes) are not significantly correlated for Rex1 and Nanog, in contrast to those from the low Stella compartment (filled boxes, bottom).

Back to article page