The Biological Basis of Cognitive and Physical Resilience

Dr. Cornelis with the assistance of Dr. Vu will identify SNPs associated with nutritional status and resilience. Multiple (non-genetic) markers of resilience will be identified by studying differences among the various measures of nutritional status and health as assessed in the UK Biobank including biomarkers of nutrition, sleep data, medical records and mental health measures. Novel SNPs associated with new measures of resilience will be identified if possible and candidate SNPs known to be associated with resilience will be examined. Dr. Cornelis and Dr. Vu will collaborate on the creation of a composite measure of resilience based on factor and/or cluster analysis of mental health and sleep variables. Resilience trajectories will be created using self-reported traumatic experience(s) as an index event using relevant data to pinpoint the date of the event(s) and then use mental health data points (baseline, touchscreen interview 2009 and the online mental health questionnaire 2016) using state-of-the-art methods such as neural networks, if appropriate, to assess resilience. Using data from the UK Biobank, genome-wide association studies (GWAS) will be conducted with resilience variables as continuous variables and ordinal variables. Accordingly, linear and logistic regression will assess relationships between measures of resilience and each genotyped/imputed SNP, adjusting for age, sex, genotyping platform and optimal principal component or cluster variables. Ancestry-stratified GWAS will be conducted. The established genome-wide threshold for significance, P<5e-8, will be used. Using summary statistics, previously reported candidates for resilience will be investigated. The significance for these hypothesis-driven tests will be defined as p=0.05/#SNPs.