From Fixed Loci To Genome-Wide, Advancements In The Detection And Analysis Of Copy Number Variants In Large-Scale Datasets

Summary
Copy Number Variations (CNVs) have been a focus of human genetic research for decades. Particular attention has been given to large recurrent CNVs, meaning deletions or duplications occurring in the same locus across different individuals in a population. Several recurrent CNVs have been associated to human disorders and syndromes, including mental and developmental disorders such as Schizophrenia Spectrum Disorder (SSD), Autism Spectrum Disorder (ASD), and Attention Deficit Hyperactivity Disorder (ADHD), often with very high risk estimates. However, such associations often originate from highly selected syndromic case studies and collections, and have been shown to not hold, at least not to the same degree, in more population representative samples such as iPSYCH2015. Moreover, recurrent CNVs represent only a minor fraction of the overall genomic variation represented by structural variants, and studies assessing the impact of CNVs on a Genome-Wide scale are still rare and unrefined, underlining how detecting and analysing CNVs across the entire genome remains a challenging task.
In this thesis, with the four main manuscripts included, and the other papers from my colleagues that I co-authored during my PhD, I present my effort to push forward the research on the biology of CNVs and their association to human traits in large SNPs-genotyped cohorts. The first manuscript describes our pipeline to reliably call and analyse recurrent CNVs. This has been the base for several other research projects. The second manuscript studies the association of NRXN1 deletions and mental disorders in the iPSYCH2015 case-cohort study. It also served as an exploratory ground for novel analytical strategies. The third and fourth manuscripts are tied together as they are centered around the genome-wide validation of CNV calls using machine vision instead of human analysts. They describe the method and its application in two large cohorts, respectively.