Connecting high-resolution 3D chromatin organization with epigenomics

While large-scale 3D genome architecture is well studied, the limits of resolution have hindered our understanding on the fine scale. Here the authors mapped 1D epigenomic profiles to fine-scale 3D chromatin structures with their deep learning model CAESAR. The model predicted fine-scale structures, such as short-range chromatin loops and stripes, that Hi-C datasets fail to detect. The resolution of chromatin conformation capture technologies keeps increasing, and the recent nucleosome resolution chromatin contact maps allow us to explore how fine-scale 3D chromatin organization is related to epigenomic states in human cells. Using publicly available Micro-C datasets, we develop a deep learning model, CAESAR, to learn a mapping function from epigenomic features to 3D chromatin organization. The model accurately predicts fine-scale structures, such as short-range chromatin loops and stripes, that Hi-C fails to detect. With existing epigenomic datasets from ENCODE and Roadmap Epigenomics Project, we successfully impute high-resolution 3D chromatin contact maps for 91 human tissues and cell lines. In the imputed high-resolution contact maps, we identify the spatial interactions between genes and their experimentally validated regulatory elements, demonstrating CAESAR's potential in coupling transcriptional regulation with 3D chromatin organization at high resolution.