Microscopy Supplement:Beyond pairwise DNA contacts: exploring higher-order genome structure using proximity ligation

Baylor College Of Medicine, Houston TX

Investigators

Linked publications & trials

Abstract

ABSTRACT A key limitation of current methods for measuring how the human genome is folded inside the nucleus of a cell is that they measure pairwise contact frequency, i.e., the likelihood that two pieces of the genome are spatially adjacent in the cell nucleus. It is well-understood that the ideal data type for the study of genome folding at a given region of the genome, would be an actual 3D trajectory showing the position of the genomic chain for every locus in that region in a single cell. Recently, the Aiden and Wu laboratories, together with collaborators, have developed a protocol for using super-resolution microscopy to `walk along chromatin', i.e., to determine the 3D position of a series of contiguous loci, and to thereby determine the actual 3D trajectory of a region of the genome. Our current methods enable us to visualize 5-10kb genomic regions with 30 (x,y) and 40 (z) nm resolution and 12 (x,y) and 50 (z) nm precision. Strikingly, we have observed strong correspondences between 3D distance, as reflected in these trajectories, and pairwise contact frequency, as measured by Hi-C. The protocol for performing such experiments is now stable, and both labs work together closely on a daily basis. Based on these findings, we believe that the super-resolution microscopy-enabled ?chromosome walking? methods that we have developed could, with additional funding, be deployed in a production setting, such as the 4D Nucleome Joint Analysis. Specifically, we propose to generate chromosome walk data spanning an entire human chromosome at 1mb and 30-40nm resolution in 100 H1ES chromosomes. We will use chromosome walk data to calibrate Hi-C and COLA data in H1ES chromosomes, translating from the absolute contact frequency of a pair of two loci to a probability distribution of pairwise 3D distances; and more generally determining the probability distribution of multi-body 3D distances conditional on a higher-order contact frequency measurement. We will also build a visualization infrastructure for chromosome walks.