Identify shared and specific interactions between diet and psychosocial and genetic factors for self-reported depression and related disorders
Approved Research ID: 1602
Approval date: June 17th 2020
Depressive illness is common and costly to the individual and society. Genetic makeup accounts for about 1/3rd of the risk of depression and environmental factors for about 2/3rds. Psychosocial adversity and stress are important aspects of the environment that contribute to depression. Other potentially important environmental factors have been little studied. It is known that what we eat and drink, our diet (carbohydrates, fats etc.) and nutrients (e.g. vitamins) is an important influence on the risk of medical disorders such as obesity, diabetes and cardiovascular disease.
These disorders are associated with an increased risk of subsequent depression and are more common in those with previous depression. This suggests that obesity-related disorders and depression may have some similar pathways of risk.
The proposed cross-sectional study aims to identify shared and specific interactions between diet and psychosocial and genetic factors for self-reported depression and related disorders. We will use the unique combination of psychosocial, dietary and mental health data available in a subset of 122,000 of the UK Biobank cohort to decisively determine whether or not there are dietary patterns and constituents that lower the risk of self-reported lifetime depression in the face of life stresses.
We will then factor-in genetic information (genotyping data will be requested, i.e. no DNA samples) and, using sophisticated statistical techniques, find new dietary and genetic factors that are highlighted because they converge on shared biochemical pathways. Understanding the role of diet in depression meets the UK Biobank's stated purpose of improved disease prevention; in contrast to genetic and psychosocial factors, dietary behaviour is potentially modifiable. For example, preventative, public health strategies could reduce the prevalence of depression by promoting resilience to psychosocial adversity and by offsetting the biochemical consequences of genetic risk.
These cross-sectional investigations identified the complex environmental- and comorbidity-network of depression, and demonstrated that comorbidity-relationships are influenced by the temporal order of diseases. Thus we would like to extend the scope of this project and carry out a longitudinal analysis in the full UK Biobank cohort. Our hypothesis is that comorbidities may inform us about the different biological pathways leading to and active in subgroups of depressed patients and thus can guide personalized medicine. In this project we aim to apply advanced machine learning methods using first occurrences of disorders and medication history to construct a temporal disease map and identify clusters of patient trajectories. Next, we will characterise these trajectories by investigating the genetic, metabolic, socio-economic, environmental, lifestyle, laboratory and cognitive profile, and calculate multimorbidity-adjusted burden of diseases. Additionally, we will perform in silico screening of multi-target drug candidates using the polypharmacy, genetic, and multimorbidity profiles of the trajectory classes. The derived results will further improve public health strategies by identifying the most frequent longitudinal multimorbidity profiles of depressed patients, and inform healthcare providers about patients' characteristics and needs, and contribute to better health risk assessment, prevention and treatment strategies.