Abstract
For over a century, scientists have been trying to understand how the DNA molecule that is so tightly packed in the micro space of the nucleus sustains critical cellular processes such as transcription, replication or the maintenance of genetic information. Despite the huge effort of researchers around the world, we know relatively little. This is in part due to the lack of methods that bring the right throughput and resolution to the study of the higher-order spatial arrangement of the genome. For example, the radial arrangement of the chromatin in mammalian cells remains largely unrevealed as well as the processes that lead to genome instability, which could potentially lead to the development of cancer. It is therefore crucial to develop tools that would allow us to precisely map both the location and frequency of DNA double-strand breaks (DSBs) along the genome, as well as to study them in the right 3D context of the chromatin. In order to fill this gap in, this thesis describes two methods which we have developed in order to map genome organization and genome fragility in the 3D space of the nucleus.
In paper I, we developed GPSeq (Genome Loci Positioning by Sequencing) as a genome-wide technique for mapping radial arrangement of the genome in mammalian cells. We showed that GPSeq accurately generates maps of the radial organization of the human genome at 1 Mb and 100 kb resolutions, thus allowing us to reveal unique radial patterns of various genetic and epigenomic traits, gene expression, A and B subcompartments as well as radial arrangements of DSBs, cancer mutations or germline variants. In paper II, we developed BLISS (Breaks Labelling In Situ and Sequencing) as a genome-wide technique to quantitatively profile DSBs distribution in cells. We demonstrated that BLISS can be successfully applied to samples with either low number of cells or to tissue sections and yet accurately detect DSBs. We showed the sensitivity of BLISS by estimating off-target activity of two nucleases- Cas9 and Cpf1 in CRISPR system and demonstrated that Cpf1 is more specific when compared to Cas9.
