Databases
In order to adequately study the complex interplay between imaging, genetics, and mental health, large datasets with deep phenotypic data are required. For such studies, researchers usually pool data in the context of a consortium such as the ENIGMA consortium1 and PGC, which focus on psychiatric genetics. Alternatively, they apply for access to large public datasets such as the abcd2 study and the UKB3, a large-scale public biomedical database from the UK which provides researchers access to a variety of data types, including genetic, imaging, demographic, clinical, and environmental data. The UKB currently contains genetic data from half a million individuals, and it aims to include imaging data from 100,000 individuals. At the time of writing, there was imaging data available for almost 67,000 participants. All work contained in this thesis relied to a large degree on data from the UKB.
A major strength of the UKB is its deep phenotyping of participants. Individuals were recruited through the British National Health Service. Participants aged 40-69 were included at an assessment centre where they filled in the consent forms, performed a self-administered questionnaire, and underwent a number of physical and functional measures (which included collection of blood, urine, and saliva)3. In addition, a number of participants were invited to participate in the imaging study, which involved a visit to one of three assessment centres where participants underwent an MRI scan. A detailed description of the genotyping and imaging acquisition and processing protocols are presented below.