3D genome architecture regulates the traffic of transcription factors throughout human chromosomes.

Transcription factors (TFs) control the rate of transcription of genetic information by binding to specific DNA sequences. The time needed for a TF to find its specific target sites is a bottleneck to the genetic response mechanism. While TF target site search is a well studied problem, the effect of genome 3D architecture on the TF target search times is poorly understood. Here, we use accurate and cell-specific 3D structural ensembles of human chromosomes to investigate how the spatial organization of binding sites on chromosomes influences the dynamics of TFs. We use Chromatin Immuno-Precipitation data to map the position of binding sites for several TFs on chromosomal structures and simulate the dynamics of individual TF within chromosomal territories. We find that the distribution of binding sites along chromosomes cooperates with the 3D folding of the chromatin fiber to induce dynamics in which TFs tend to visit sites distributed sequentially along the genome. In this way, genome 3D architecture appears to reduce the time each TF spends in the unbound state while commuting from one target site to the other. At the same time, genome 3D architecture further reduces the flux of TFs between binding sites already well separated along the genome, effectively isolating distant clusters of binding sites. We compare the TF traffic patterns generated by the 3D structures of human chromosomes with those generated by several alternative structural models characterized by increasing randomness. Finally, we study the effect of lengthwise compaction and phase separation, known architectural features of the human genome, in TFs target search. In short, our analysis demonstrates that genome architecture regulates the traffic of TFs within chromosomal territories and reduces the time each TF spends commuting between binding sites.