Figure 3

Brg1 but not BAF53a is required for the regulation of replication domains. (A) Replication timing profiles of mock- and 4-hydroxytamoxifen (OHT)-treated Brg1 flox/flox and BAF53a flox/-, displayed as in Figure 1B. Plots for the average of two biological replicate experiments are shown. (B) Correlation analysis between replication timing datasets (Pearson’s R values). (C) Correlation analysis of replication timing differences (dRT). Replication timing data were averaged into 200-Kb windows and dRTs (that is, Brg1 OHT ratio - Brg1 mock ratio) were calculated for comparisons between different groups. A substantially lower conservation of replication timing affected regions is seen between BAF53a ESCs and BAF250a or Brg1 embryonic stem cells (ESCs) than between BAF250a and Brg1 ESCs. (D) Summary of significant early to late (EtoL) and late to early (LtoE) switching segments in OHT-treated cells as determined in Figure 1C. (E) Box plots show the replication timing of segments that are sensitive to BAF250a loss (false discovery rate (FDR) = 1% domains identified in Figure 2G) in various esBAF subunit mutants. Replication timing in mutants (red) is compared to that in control counterparts (green). P values were calculated using the Wilcoxon rank sum test. (F) Replication timing plots of exemplary EtoL switching domains. These domains undergo EtoL switching in BAF250a and Brg1 mutants, but not in BAF53a mutant.