Decoding chromosome organization using proximity labeling and long-read sequencing

Genomic approaches have provided detailed insight into DNA-protein interactions and chromosome architecture. However, commonly deployed techniques do not preserve connectivity-based information, leaving large-scale genome organization poorly characterized. Here, we develop npDamID, an in vivo proximity-labeling technique that indelibly marks, and then decodes, protein-associated sites. npDamID tethers dam methyltransferase to a protein of interest, followed by Nanopore sequencing to identify methylated bases along reads >100 kb. As proof-of-concept, we analyze, in budding yeast, a conserved cohesin-based meiotic backbone that organizes chromatin into an array of loops. Our data recapitulate the pertinent features of known cohesin association patterns and, importantly, expose variability between cells. Analysis of single reads reveals distance-dependent short- and long-range correlation between adjacent methylated bases. Finally, an important advantage of our approach is the ability to define protein association patterns in repetitive regions. By anchoring ultra-long reads onto unique regions, we define in vivo cohesin association patterns within the ribosomal DNA locus. Our versatile technique promises to illuminate diverse chromosomal processes by providing a cumulative record of heterogeneous association patterns of chromosomal factors as well as the in vivo conformations of single chromosomes.