Fig. 4

RPA local density at individual replication forks estimated by the TC function. a–c Representative U2OS nucleus labeled with RPA70 (R), PCNA (B), and MCM (G) were treated with 0, 0.2, and 0.6 μM Aphidicolin (APH), respectively. Scale bar = 2500 nm. d–f Configurations resolved from N = 80, 95, and 101 nuclei treated with 0, 0.2, and 0.6 μM APH, respectively (from 3 experimental replicates). Filtered by the criteria Dist_BG (distance between PCNA and MCM) ≤ 100 nm, 25, 21, and 26 configurations from [APH] = 0, 0.2, and 0.6 μM, respectively, were overlaid by aligning the BG edge onto the same horizontal, so that the relative distribution of PCNA (B) to RPA70 (R) and MCM (G) is clearly seen to form a fork-like pattern where RPA were mostly found in between the PCNA and MCM. The circle size at each vertex represents the local density of RPA at each individual fork as calculated by multiplying the conditional probability of finding RPA with each given pairwise PCNA-MCM and the overall average density of RPA. g Statistics of the average local density of RPA at each individual fork (as displayed as the size of the circles in d–f). Each plot displays such average local density from one nucleus. p = 0.043 and 0.036 for unpaired t-test between [APH] = 0 vs. [APH] = 0.2 μM and [APH] = 0 vs [APH] = 0.6 μM, respectively (*, unpaired t test, 0.01 ≤ p < 0.05). h, i A molecular model of RPA accumulating in between PCNA and MCM in non-treated control (Normal Replication, NT) cells (h) and APH stressed (Replication Stress, APH) cells (i). Source data for Fig. 4d-4j are provided as a Source Data file