Extended Data Fig. 4: MN-seq characterization and micronucleus content bias in nocodazole-treated and cancer cells.

a. Live-cell imaging of RPE1-hTERT cells to determine survival when treated for 16 h with 62.5 nM Cpd-5. Experiment was performed in triplicate (mean ± s.e.m., two-tailed Fisher’s exact test). b. Representative images of a FAC-sorted nucleus and micronucleus (scale bar, 5 μm). c. Gating strategy for micronuclei. d,e, Representative FISH images (d) and quantification (e) of micronucleated RPE1-hTERT cells treated with Cpd-5 for 16 h (scale bar, 5 μm). The graph shows the percentage of micronuclei containing a certain chromosome (s.e.m., two-tailed Pearson correlation coefficient, 1/15 (n = 371 micronuclei), 2/17 (n = 277 micronuclei) and 6/18 (n = 376 micronuclei)). Three independent experiments were performed. f,g,h, as in Fig. 2b–d but instead cells were treated with low nocodazole (single micronuclei; mean ± s.e.m, two independent experiments, n = 151, bulk micronuclei; mean ± s.e.m., four independent experiments, two-tailed Pearson correlation coefficient, n = ~8000 micronuclei). i,j,k, Log2-fold enrichment of chromosomes in MN determined from bulk MN-seq data sorted from three chromosomally unstable cancer cell lines (mean ± s.e.m., two-tailed Pearson correlation coefficient, n = ~1500 MN for HT-29 and WiDr and 6000 for U2OS). Three independent experiments were performed for each cell line.