Depression is the most common psychiatric disorder and the leading cause of disability worldwide. Despite evidence for a genetic component, the genetic aetiology of this disorder remains elusive. To date, only one association study has identified and replicated risk loci for depression. This thesis focuses on aiding genetic discovery by revisiting the depressed phenotype and developing a quantitative trait, using data from Generation Scotland: The Scottish Family Health Study. These analyses aim to test whether this derived quantitative trait has improved statistical power to identify genetic risk variants for depression, relative to the binary classification of case/control. Measures of genetic covariation were used to evaluate and rank ten measures of mood, personality and cognitive ability as endophenotypes for depression. The highest ranking traits were subjected to principal component analysis, and the first principal component used as a quantitative measure of depression. This composite trait was compared to the binary classification of depression in terms of ability to identify risk loci in a genome-wide association study, and phenotypic variance explained by polygenic profile scores for psychiatric disorders. I also compared the composite trait to the univariate traits in terms of their ability to fulfill the endophenotype criteria as described by Gottesman and Gould, namely: being heritable, genetically and phenotypically correlated with depression, state independent, co-segregating with illness in families, and observed at a higher rate in unaffected relatives than in unrelated controls. Four out of ten traits fulfilled most endophenotype criteria, however, only two traits - neuroticism and the general health questionnaire (a measure of current psychological distress) - consistently ranked highest across all analyses. As such, three composite traits were derived incorporating two, three, or four traits. Association analyses of binary depression, univariate traits and composite traits yielded no genome-wide significant results, with most traits performing equivalently. However, composite traits were more heritable and more highly correlated with depression than their constituent traits, suggesting that analyzing these traits in combination was capturing more of the heritable component of depression. Polygenic scores for psychiatric disorders explained more trait variance for the composite traits than the univariate traits, and depression itself. Overall, whilst the composite traits generally obtained more significant results, they did not identify any further insight into the genetic aetiology of depression. This work therefore provides support for the urgent need to redefine the depressed phenotype based on objective and quantitative measures. This is essential for risk stratification, better diagnoses, novel target identification and improved treatment.
| Identifer | oai:union.ndltd.org:bl.uk/oai:ethos.bl.uk:738710 |
| Date | January 2017 |
| Creators | Hall, Lynsey Sylvia |
| Contributors | McIntosh, Andrew ; Sibley, Heather ; Haley, Christopher |
| Publisher | University of Edinburgh |
| Source Sets | Ethos UK |
| Detected Language | English |
| Type | Electronic Thesis or Dissertation |
| Source | http://hdl.handle.net/1842/28737 |
Page generated in 0.0016 seconds