Extended Data Fig. 10: Functional effects of nonsense variants in relation to position and stop codon context.

a, Function scores are plotted by position in exon 1 of VHL and colored to highlight nonsense variants. All nonsense SNVs tested between c.160A and c.601C scored as depleted, except for c.264G>A, a variant associated with type 2 VHL disease²⁴. b,c, Function scores for n = 40 nonsense variants between c.160 and c.601 are plotted by termination codon (b) and 4-bp termination codon context (c). Differences between function scores by termination codon were tested using a one-way ANOVA. (Boxplot: center line, median; box limits, upper and lower quartiles; whiskers, 1.5× interquartile range; all points shown.) In c, the blue line indicates the mean score for each stop codon context and the size of each dot corresponds to the number of cBioPortal ccRCC samples in which the SNV has been observed. d, A dual-fluorophore stop-codon readthrough (SCR) reporter assay was used to quantify readthrough of nonsense variants assayed by SGE. Nonsense variants with 138 bp of surrounding VHL sequence were cloned between EGFP and mCherry, such that mCherry expression only occurs if the nonsense codon fails to terminate translation. e,f, Flow cytometry data for live populations of single cells are shown for each plasmid tested, with gating to determine the fraction of transfected cells (EGFP⁺) positive for mCherry expression. Data were normalized to a control vector without a stop codon (pSCR-VHL-no-stop). Control plasmids are in (e), and plasmids containing VHL nonsense codons are in (f).