The code we share: Leveraging genomic diversity to understand disease

The aim of this thesis was to add to the elucidation of genetic variants that contribute to human disease. This was done by using conventional methods, such as by investigating founder variants (variants that are enriched within an isolated population) that cause infertility and cardiovascular disease, but also by using recently developed methods such as genomic structural equation modelling (SEM). In chapter 2, we found that biallelic (likely) pathogenic variants in MSH4 and MSH5 lead to azoospermia, caused by meiotic arrest. In chapter 3, we learned that a homozygous loss-of-function variant in cardiac transcription factor HEY2 leads to critical congenital heart disease, and that rare, likely pathogenic heterozygous variants in HEY2 functional domains are associated with cardiovascular defects. Moving from clear founder variants to more subtle effects of parental relatedness on human diseases, in chapter 4 we investigated the association between autozygosity (the proportion of the genome that is in runs of homozygosity, which reflects the degree of parental relatedness) and the full spectrum of human diseases. We observed positive associations between autozygosity and the total number of disease subchapters with at least one diagnosis and those within respiratory and endocrine chapters, especially type II diabetes. Finally, using genomic SEM, we studied the genetic structure underlying metabolic syndrome in chapter 5. We revealed 235 genomic risk loci which harbour genes that are, amongst others, involved in lipid remodelling and show increased expression in the brain, especially in dopaminergic and GABAergic neurons.