Fig. 3: Genomic landscape of SSBs.

a Average fraction of SSBs falling within each of the indicated genomic elements for different biological replicas of K562 drug treatments (B1 or B2), no-formaldehyde crosslinking (“no-form”), PBMCs and detected using either SSiNGLe-SMS (combined K562 or PBMC data) or SSiNGLe-ILM. b Box plots of distributions of odds ratios of overlap between different genomic elements and different sample types. The top panel shows results from SSiNGLe-SMS for K562 (left box plot) and PBMCs (right box plot). The bottom one presents the results from SSiNGLe-ILM with four box plots per element representing K562 B1, K562 B2, K562 no-formaldehyde and PBMCs (left to right). c Distribution of odds ratios of overlap between different genomic elements (Y-axes) and drug treatment times (X-axes) for two biological replicas (B1 and B2) and no-formaldehyde crosslinking K562 samples (SSiNGLe-ILM). d Box plots of distributions of template/non-template ratios for exons (left box plot) and introns (right box plot) for different sample types. SSiNGLe-SMS (left panel) and SSiNGLe-ILM (right panel). e Box plots of distributions of enrichment of SSBs for the indicated sample types in different phases of the cell cycle (G1, S1–S4 and G2). For each phase and sample types, data from all SSBs (left box plot) and those overlapping common SNPs (right box plot) are shown. The results were obtained with SSiNGLe-ILM. f Box plots of distribution of odds ratios of overlap between SSBs and SNPs or indels (common, low frequency and rare) for the indicated distance windows. Top and bottom panels represent respectively SSiNGLe-SMS and SSiNGLe-ILM. For each variant type and distance combination, the top panel shows results from SSiNGLe-SMS for K562 (left box plot) and PBMCs (right box plot), and the bottom panel shows the results from SSiNGLe-ILM for K562 B1, K562 B2, K562 no-formaldehyde and PBMCs (left to right). g Box plots of distribution of nucleosome occupancy scores for the indicated sample types and whole genome (SSiNGLe-ILM). h Decrease of the complexity of SSBs (unique breaks/ total filtered reads, Y-axis) vs increase in the total number of filtered reads (X-axis) in the deep sequencing samples from indicated treatments (SSiNGLe-ILM).