Skip to main content
Fig. 4 | BMC Biology

Fig. 4

From: Migration through a small pore disrupts inactive chromatin organization in neutrophil-like cells

Fig. 4

Short-range contacts were depleted after migration with constriction. a The whole genome contact matrix at 100 kb resolution was normalized with iterative correction and eigenvector decomposition (ICE), and distance-dependent contact probability plotted on a log-log scale. There was no significant difference between the distributions of unmigrated cells, and cells migrated without constriction (14-μm pores) (KS test, p = 0.14), while cells migrated with constriction (5-μm pores) have a significantly different distribution from both unmigrated and migrated without constriction (KS test 5.74 × 10− 9 and 3.11 × 10− 13 respectively). The inset highlights the rapid decay of contact frequency between 100 kb and 500 kb in migration with constriction. b Differential intrachromosomal contacts between migration with and without constriction were called at 100 kb resolution, tiling across the chromosome in 40-kb bins. Long-range (> 1 Mb) differential contacts were equally likely to be lost or gained, while short-range (100 kb–1 Mb) contacts were significantly more likely to be lost after migration with constriction. Regions involved in a contact were assigned to compartment A or B. c Distribution of disrupted contacts. Bin sizes are uneven. i Number of significant contacts (FDR < 0.05) in compartment A and B in cells migrated without constriction. ii Number of significantly decreased contacts (FDR < 0.1) in compartments A and B between migration with and without constriction. iii Bootstrapping of significant contacts found a strong enrichment of decreased short-range (< 1 Mb) contacts in compartment B. Error bars represent 99% CI of the expected proportion of contacts in compartment A or B based on 10,000 bootstraps. d 100-kb bins were defined as transcriptionally active if they contained one of more expressed gene. i Stable A compartment regions were most likely to be active, and stable B compartment genes were least likely to be active. ii A two sample test for equality of proportions found that disrupted contacts in either compartment A or B were less transcriptionally active than total contacts in the same compartment. Error bars show the 95% confidence interval of the difference in proportion of disrupted and total contacts that contain at least one expressed gene

Back to article page