Polymer modelling to investigate yeast 3D chromosome organization in S-phase and mitosis
Eukaryotic genomes are hierarchically organized into loops, TADs, spatial compartments and chromosome territories. Characterizing the formation and function of such non-random 3D organization constitutes a major challenge in modern biology. For example, experimental evidence suggests that the DNA replication process is coupled to genome folding: replicating forks may arguably colocalize spatially forming the so-called replication factories, while TADs boundaries have been associated with replication origins and A/B compartmentalization with domains of different replication timing. While the molecular actors and mechanisms involved in DNA replication on one side and in genome folding on the other start to be well characterized independently, the interplay between the two remains elusive. To start addressing this question, we focused on how chromatin folding may evolve as a consequence of the mechanical stress induced by genome duplication. In particular, we developed an original polymer model capable of self-replicating itself starting from several origins of replication. The developed computational framework allows us to simultaneously describe 3D chromatin folding and 1D replication dynamics and to address how specific molecular processes may influence the genome organization. In particular, we contextualized the model to the S-phase of Saccharomyces cerevisiae. From our simulations, we predicted two possible mechanisms that may drive spatial organization around origins of replication: an entropic repulsion due to the ring topology of replication bubbles and an active extrusion mechanism driven by putative co-localization of sister forks, leading to the formation of ‘chromatin fountains’ around origins for which we find a signature in experimental Hi-C data. Our model also enabled us to investigate how the establishment of the cohesion between sister chromatids and the mitotic condensation are connected with the replication process. In particular, our results show how simply starting from the ChIP-Seq profile of cohesin and tuning a minimal set of free parameters, it is possible to recover some structural properties of yeast mitotic chromosome observed experimentally, leading to a better understanding of the interplay between cohesion, mitotic condensation and replication.