Polymer modelling accurately predicts three-dimensional chromosome reorganization with a seasonal coronavirus infection

Many vital processes, such as DNA transcription, replication, and repair, depending on the cell's structural organization of chromosomes. Three-dimensional genome spatial information is essential for understanding gene regulation and epigenetic effects, which can regulate the 3D architecture of the whole genome. Various advances in experimental techniques such as chromosome conformation capture (Hi-C), Assay for Transposase-Accessible Chromatin with high-throughput sequencing (ATAC-seq) and other related methods have led to the finding of topologically associating domains (TADs), chromatin loops, and different compartments existing in the chromosomes. However, less progress has been made in modelling chromosomes based on experimentally-driven approaches. Previously, polymer simulations were employed, incorporating one-dimension information and generating connectivity profiles for the chromosomes. Our method (4DHiC) extracts 2D information from their corresponding Hi-C maps and implements it as a harmonic constraint in the polymer model. Here, different Hi-C maps for MRC cell lines infected with and without seasonal coronavirus infection at varying stages of infection were generated to test the model's applicability. We performed Langevin dynamics to simulate the polymer model with HiC-map constraints using the 4DHiC method. On comparing the simulated and experimental HiC map, a strong correlation was observed (point-to-point RPearson∼0.75). Along with the model development, this study also revealed the critical structural difference that arises from viral infection in the chromosome organization. One important chromosome organization feature, A/B compartmentalization, was also studied. A clear separation was observed based on the principal component analysis, predicting the change in compartment contacts. Another advantage of the method is the scalability, i.e. can vary the number of beads based on the resolution of the HiC map. Lastly, we emphasize that the protocol is general and can be used to model the whole genome or individual chromosomes for any system.