Skip to main content
. 2022 Sep 26;13:5632. doi: 10.1038/s41467-022-33290-0

Fig. 4. Intra-tumour RAS84 heterogeneity.

Fig. 4

a Heatmap showing RAS84 gene expression across the TRACERx multi-region cohort. The samples have been classified into the five RAGs using an SVM classifier. Contiguous samples from the same patient are indicated by the grey shades just at the top of the heatmap. Gene mutation status is indicated at the top, KRAS in red, EGFR and TP53 in grey. Per-patient regions are indicated by the rows at the bottom of the heatmap. KRAS mutants are shown in red, wild-type in green. Patients with regions spanning multiple RAGs are indicated with an asterisk. Identifiers are given on the left in the three cases where regions did not fall into adjacent RAGs. b Intra- (orange) and inter-tumour (grey) sample Euclidian distant distributions. The maximum intra-patient distance for patients with samples spanning different RAGs are indicated with an asterisk. c Plot showing the enrichment of RAS84 genes in genes previously classified as low for intra-tumour expression variance and high for inter-tumour expression variance in the TRACERx lung adenocarcinoma cohort. This group is represented by quadrant 4 on the plot showing a 3.2x fold enrichment of RAS84 genes relative to all expressed genes (Fisher’s exact enrichment p-value 9.06e−6).