Fig. 4: Role of TP53, cell-cycle and replication-associated genes in proximity bias.
Box and scatter plots of whole-genome-level proximity bias quantification by one-sided Brunner–Munzel intra-arm versus inter-arm probability from DepMap 22Q4 data with the cell lines stratified by gene status. Each point represents a bootstrap sample of cell lines; 128 bootstraps were run for each condition. The box plots show the median and lower and upper quartile with whiskers extending to the furthest points within 1.5 times the interquartile range. The P values for two-sided t-tests with Bonferroni correction applied are given, and the detailed test statistics are given in Supplementary Table 3. Horizontal dashed lines at 0.5 indicate the expected baseline if no proximity bias effect was present. a, Comparison between cell lines with WT TP53 and those with TP53 LOF (n = 266 and 277 cell lines, respectively, 212 sampled in each bootstrap; P < 1 × 10−10). b, Selected genes and conditions LOF or AMP in TP53 WT background (CDKN2A: WT n = 112, LOF n = 136, P < 1×10−10; CDKN2B: WT n = 124, LOF n = 132, P < 1 × 10−10; BTG2: WT n = 173, AMP n = 87, P = 1 × 10−6; MDM2: WT n = 210, AMP n = 52, P < 1 × 10−10; MDM4: WT n = 174, AMP n = 87, P = 2 × 10−10; n = 20 cell lines were sampled in each bootstrap for all genes). c, Selected genes and conditions in TP53 LOF background (CDKN2A: WT n = 82, LOF n = 156, P < 1 × 10−10; CDKN2B: WT n = 103, LOF n = 154, P < 1 × 10−10; BTG2: WT n = 188, AMP n = 81, P = 5 × 10−10; MDM2: WT n = 218, AMP n = 43, P = 1; MDM4: WT n = 187, AMP n = 79, P = 6 × 10−9; n = 20 cell lines were sampled in each bootstrap for all genes). The genes with fewer than 25 cell lines in a given condition are not shown.
