Skip to main content
. 2019 Nov 28;9:17754. doi: 10.1038/s41598-019-54435-0

Figure 6.

Figure 6

Model describing the emergence of divergent subtypes of SCC. (a) Illustration of the proposed state of samples according to the distribution in PC1-PC2 dimensions. Normal: skin. (b) Immune cell infiltration scores and (c) Average lesion thickness re-grouped according to quadrant locations. (c) Kruskal-Wallis test (P = 0.0007) followed by Dunn’s multiple comparisons test. (d) Upper-left, (e) upper-right and (f) lower-right scores (logit; log-odds) for logistic regression analysis for samples in each diagnosis category. One-way ANOVA with Tukey’s multiple correction was performed where **P < 0.01; ****P < 0.0001. (g) X-Y plot of upper-right scores versus upper-left scores for AK showing negative correlation for samples scoring highly in either signature.