Fig. 5

SSVs associated with TAD disruption, enhancer hijacking, and rearrangement of regions with high or low methylation. a As compared to all SSVs, fractions of SSVs involving topologically associated domain (TAD) disruption and altered gene expression or DNA methylation (defined as FDR < 5% for the gene or CGI probe within the given region window, with corrections for cancer type and CNA, and expression > 0.4SD or < − 4SD from median for the case harboring the breakpoint). p values by chi-squared test. b Percentages of SSV breakpoint associations involving the translocation of an active, in vivo-transcribed enhancer [28] within 0.5 Mb of the gene (where the unaltered gene had no enhancer within 1 Mb), as tabulated for the entire set of SSV breakpoint associations occurring 0–500 kb upstream of a gene and with breakpoint mate on the distal side from the gene, as well as for the subsets of SSV breakpoint associations involving altered gene expression or CGI methylation (using FDR < 5% by distance metric and > 0.4SD or < − 4SD, with corrections for cancer type and CNA). p values for enrichment as compared to all SSV events by chi-squared test. c Using a dataset of DNA methylation of normal tissues (Additional file 1: Figure S7), the average DNA methylation represented by the rearranged region (using window of 50 kb) was compared with that of the CGI nearby the gene on the other side of the SSV breakpoint, with the average difference in methylation beta values computed for all SSV-CGI associations, as well as for the subset of SSV-CGI associations involving higher or lower DNA methylation (defined as for b). p values by Spearman’s rank correlation. Bars represent standard error. d By gene and by cancer type, the number cancer cases involving the rearrangement of a region of low methylation (average methylation beta difference < − 0.1), with corresponding decrease in methylation and increase in expression being observed (< − 4SD and > 0.4SD from median, respectively), involving 41 genes and 105 cases (genes affected in > 2 cases or cancer-associated genes [14,15,16] being represented here). See also Additional file 1: Figure S7 and Additional files 8, 9, and 10