Defining the phenotype: Combining phenotypes for discovery of novel genetic associations

Our genomes are 99.5% identical, and yet we are different. Part of this variation can be explained by genetic variation. The genomic locations where a certain nucleotide (a different 'letter' in the DNA sequence) is associated with a phenotype can be found with a genome-wide association study (GWAS): we compare - at genomic locations with a lot of variation (single nucleotide polymorphisms; SNPs) - people with and without the phenotype (cases respectively controls). In this thesis I show the added value of defining the phenotype differently.


In chapter 1 we look for genetic associations with the subtypes of ischemic stroke. Three existing methods to diagnose the subtype do not always agree with each other. We compared these methods, and the intersect and union. The intersect (where cases were diagnosed with the subtype by all methods) is a stricter definition that only counts people as a case if they quite certainly have a subtype. In the case of ischemic stroke it seems more important to know the subtype with high certainty than it is to have a large sample.


In chapter 2 we describe a GWAS of the age at onset of ischemic stroke. We split the cases in women and men, and found a SNP that is associated with a 1.6 years lower age at onset of ischemic stroke in women.


The last chapters are on pleiotropic SNPs; they have an association with multiple phenotypes. Chapter 3 describes methods that can find pleiotropic SNPs. Chapter 4 describes a method we have developed; PolarMorphism. We use a different approach than previously described methods. PolarMorphism is faster then the method that previously could find the most pleiotropic SNPs, without finding less SNPs.