Abstract
Pathological forms of mood have been documented for centuries. Mood disorder symptomatology varies enormously between individuals and can be viewed as a spectrum from depression to mania. The two most common disorders within the mood spectrum are major depressive disorder and bipolar disorder. The precise mechanisms that give rise to these disorders are largely unknown, but we now understand that both genetic and environmental factors increase individual risk. Improving knowledge of risk factors for mood disorders is crucial because they are leading drivers of disability and can substantially impact the quality of life of the affected individual.Recent advances in technologies aimed at mapping people’s inherited DNA have facilitated a deeper exploration of the genetic basis of mood disorders. Genome-wide association studies, which utilise genotype data, have been pivotal in confirming mood disorders’ polygenic, heritable nature, as well as demonstrating that genetic risk factors are shared between psychiatric disorders. Genome-wide association studies of complex traits/diseases require sample sizes in the thousands to effectively capture the small effects of individual genetic variants. This is an even bigger priority for mood disorders due to their highly polygenic genetic architectures. Another factor is the level of detail included in the mood disorder phenotype, partly because this dictates sample size, but also because trait heterogeneity influences statistical power.
Research studies and biobanks that collect self-reported data on participants’ psychiatric health, in addition to DNA samples, have facilitated the cumulation of samples sufficient for genetic studies of mood disorders. The UK Biobank, the Genetic Links to Anxiety and Depression study, and the COVID-19 Psychiatry and Neurological Genetics study are three UK-based studies that offer an opportunity to apply statistical genetics methods to self-reported data on disorders and symptoms within the mood spectrum. Effectively studying the genetics of any trait is rooted in the validity of the way it is measured, and there are many possible modes of measuring the mood spectrum with self-reported data. Given the sharp rise in genetic studies of mood disorders and the growing acceptance of their heritability, it is timely to evaluate approaches to measuring this disorder spectrum, to increase statistical power and to maximise the chance of replicable findings.
In this thesis, three empirical chapters are presented that explore three approaches and evaluate their utility for genetics research. The three approaches are: diagnostic subtypes, continuous measures, and analyses at the symptom-level (including symptom subgroups and individual symptoms). The first empirical study (chapter 2) focuses on refined phenotyping approaches to explore the relationship between self-reported trauma and major depressive disorder. This chapter calculates the genetic overlap between various subtypes of this mood disorder and posttraumatic stress disorder to examine whether they share a genetic basis for trauma sensitivity. The second empirical study (chapter 3) investigates whether the Mood Disorder Questionnaire, a widely used screening tool for bipolar disorder, can be leveraged to construct a continuous measure of mania. This chapter examines whether this mania phenotype is valid for genome-wide association studies. The final empirical chapter (chapter 4) also applies a continuous measure to a specific mood symptom: anhedonia. Anhedonia has been posited as a risk factor for treatment resistance in individuals with major depressive disorder. This chapter examines whether the two mood phenotypes share genetic risk factors.
Genome-wide association studies hold great promise for improving the lives of individuals affected by mood disorders. However, the quality of their findings depends on the quality of the phenotypes examined. The final chapter (chapter 5) draws conclusions from the three studies together, and comments on the lessons learnt for phenotyping the mood spectrum for genetic studies. The thesis finds that the heterogeneity of mood disorders and their symptomatology can be accurately captured through various types of self-reported data and phenotyping strategies, but there are important caveats to this. An evaluation of the suitability of clinical tools when used for self-reported data collection should be prioritised. Also, data incorporated into continuous measures of mood psychopathology, such as composite symptom scores or staging models, requires careful consideration to reduce phenotypic noise and maximise statistical power. The hope is that the lessons presented in the thesis will be useful for other researchers who endeavour to study the genetic basis of mood disorders in the future.
| Date of Award | 1 May 2023 |
|---|---|
| Original language | English |
| Awarding Institution |
|
| Supervisor | Robin Murray (Supervisor) & Gerome Breen (Supervisor) |