Study of complex chromosomal rearrangements in cancer. The role of extrachromosomal circular DNA as a genome remodeler in neuroblastoma

Abstract

[eng] This thesis illustrates the work I have developed as a Ph.D. student in the computational genomics group lead by Dr. David Torrents at the Barcelona Supercomputing Center. The group’s expertise in the analysis of biological data and the detection of variants to gain more knowledge about the genetic and molecular implications of human diseases, such as cancer, has allowed me to learn and conduct my research.

Focusing on the analysis of structural variation in cancer, I have been able to apply different methodologies for sequencing data, retrieving, filtering, and determining the mutational profile foreach ofthe studied samples.Moreover,I have characterized new patterns of genomic rearrangements related to transposase-derived genes and extrachromosomal circular DNA elements in cancer. Therefore, this thesis is centered in the study of the genomic variation and mechanisms associated with oncogenic processes together with the analysis of elements of the human genome that are not generally included in comprehensive cancer studies, such as circular DNA elements.

In summary, starting with the introduction, I give an overview of the methodological aspects of the study of cancer through the impact of sequencing technologies, the biological and molecular causes and consequences of this disease, focusing on structural variation, and the description of the circular DNA genomic component and its known implications in cancer. Finally, I introduce neuroblastoma, an example of how structural variants and circular DNA drive tumorigenesis.

Next, I present the results of this thesis in three blocks, all of which have in common the study of structural variation in cancer. Two of the blocks correspond to the PGBD5 and neuroblastoma publications, and one corresponds to the continuation of the PGBD5 analysis in ICGC-Pan-Cancer data.

As an overview of the trajectory of this thesis, I started with my involvement in a project focused on analyzing the role of PGBD5 —a transposase-derived gene—

as an oncogenic mutator with an associated mechanism for site-specific DNA rearrangements. In this study, we describe how the expression of this gene promotes cell transformation and the generation of recurrent rearrangements, presenting a conserved motif in cell lines and childhood tumors. As a logical continuation of this publication and thanks to the access of our group to ICGC-PCAWG data, we expanded the study of these characteristic PGBD5-motif-related rearrangements to different patients and tumor types.

The following part of this thesis is focused on the analysis, description, and classification of the genomic somatic rearrangements in neuroblastoma. With the aim of better grouping the patients with different clinical outcomes, we searched for differential patterns of structural variants across the samples. From this analysis, we were able to describe a new phenomenon that connects circular DNA with different integration sites around the genome through complex rearrangement clusters providing evidence on how circular DNA can act as a driver of genomic remodeling in neuroblastoma.

To finalize, I present the general discussion of the results and questions addressed in this work to, then, end up disclosing the final conclusions of this thesis.