Extended Data Fig. 2: CRISPR-addiction scanning reveals drug addiction mutations.
From: Drug addiction unveils a repressive methylation ceiling in EZH2-mutant lymphoma

a, Bar plot of wild-type (gray) and CXCdel (red) allele frequencies (y-axis) from CRISPR-addiction scanning. Data are represented by mean of n = 3. b, Histogram showing distribution of addiction scores (x-axis) among the three replicates of the GSK343 (left) and EED226 (right) CRISPR-addiction scanning experiments. c, Graph showing proportions of each sgRNA (y-axis) in each replicate of the GSK343 CRISPR-addiction scan over time (x-axis). Curves were calculated using the estimated intrinsic growth rate r for each sgRNA-containing cell population under the assumption of competitive logistic growth. The reference sgRNA is shown in black, sgRNAs demonstrating pure resistance are in blue and addicted sgRNAs are in red. d, Same as in c. but for each replicate of the EED226 CRISPR-addiction scan. e, Plots showing simulated densities for three hypothetical sgRNA populations. sgRNA densities begin at 1/500. The intrinsic fitness (uninhibited growth rates) for sgRNA populations 1–3 are 1.51, 1.50, and 1.49, respectively. When inhibitor is removed, the fitness of sgRNA populations 1 and 2 remain unchanged, while the fitness of sgRNA 3 decreases to 1.20. The sgRNA densities over time are described by Equation 7 (see Supplementary Note). The total density is equal to the sum of the densities of the three sgRNAs and is described by Equation 2. f, Plots showing simulated sgRNA proportions according to Equation 7. Trajectories are derived by dividing the sgRNA densities depicted in e. by the total population size. g, Plots showing simulated sgRNA proportions according to Equation 9. sgRNA proportions start at 0.33. The intrinsic fitness (uninhibited growth rates) for sgRNA populations 1–3 are 1.51, 1.50, and 1.49, respectively. When inhibitor is removed, the fitness of sgRNAs 1 and 2 remain unchanged, while the fitness of sgRNA 3 decreases to 1.20. h, Plots showing Z-values for simulated sgRNAs relative to sgRNA population 1. The definition of Z is given in Equation 10. The slopes of these lines give may be used to compute the change in intrinsic fitness for each sgRNA when on drug vs. off drug. Calculation of addiction scores is detailed in Supplementary Note.