Speaker
Description
The rapid advancement of comprehensive genome mapping techniques, such as Hi-C, for investigating the three-dimensional configuration of the genome within the nucleus has uncovered complex chromatin architectures at multiple scales, including A/B compartments, topologically associating domains (TADs), and chromatin loops. These structural elements of the 3D genome are linked to crucial genomic functions, such as gene transcription, although the variability of 3D genome structures and their functional implications at the single-cell level remain largely elusive. Emerging single-cell Hi-C (scHi-C) technologies now facilitate the genomic mapping of 3D chromatin configurations in individual cells, offering the potential to elucidate fundamental connections between genome structure and function at single-cell resolution across diverse biological contexts. Nevertheless, there is a significant deficiency in computational methodologies capable of physically characterizing the sparse scHi-C data. Here, we employ a polymer-physics based approach, which relies on phase-separation mechanisms, combined with machine learning, to impute contact maps from the sparse scHi-C data and to analyze the cell-to-cell variability of three-dimensional (3D) chromatin organization through their polymer models.
