Fig. 2

SNVs generated during long-term niraparib treatment. a The mean number of SNVs generated per sequenced genome in each indicated cell line following mock treatment or treatment with the indicated concentration of niraparib. Red symbols show the values for individual samples, error bars indicate SEM. b–e Mean triplet SNV mutation spectrum of the mock treatment (top panel) or niraparib treatment (bottom panel) in the indicated cell lines. Each mutation class, as indicated at the top of the panel, is separated into 16 categories based on the identity of the preceding and following nucleotide as shown below. The order of the following nucleotides, not shown due to lack of space, is alphabetical. f Similarities and differences of the triplet SNV spectra of individual samples visualised using t-distributed stochastic neighbour embedding (t-SNE). g Comparison of the mean triplet SNV spectra shown in b–e to COSMIC mutation signatures using t-SNE. Experimental spectra are labelled according to the key shown on the right; COSMIC signatures are numbered