Major depressive disorder (MDD) has been the subject of many neuroimaging case control classification studies. Although some studies report accuracies =80%, most have investigated relatively small samples of clinically-ascertained, currently symptomatic cases, and did not attempt replication in larger samples. We here first aimed to replicate previously reported classification accuracies in a small, well-phenotyped community-based group of current MDD cases with clinical interview-based diagnoses (from STratifying Resilience and Depression Longitudinally cohort, STRADL ). We performed a set of exploratory predictive classification analyses with measures related to brain morphometry and white matter integrity. We applied three classifier types SVM, penalised logistic regression or decision tree either with or without optimisation, and with or without feature selection. We then determined whether similar accuracies could be replicated in a larger independent population-based sample with self-reported current depression (UK Biobank cohort). Additional analyses extended to lifetime MDD diagnoses remitted MDD in STRADL, and lifetime-experienced MDD in UK Biobank. The highest cross-validation accuracy (75%) was achieved in the initial current MDD sample with a decision tree classifier and cortical surface area features. The most frequently selected decision tree split variables included surface areas of bilateral caudal anterior cingulate, left lingual gyrus, left superior frontal, right precentral and paracentral regions. High accuracy was not achieved in the larger samples with self-reported current depression (53.73%), with remitted MDD (57.48%), or with lifetime-experienced MDD (52.68 60.29%). Our results indicate that high predictive classification accuracies may not immediately translate to larger samples with broader criteria for depression, and may not be robust across different classification approaches.
STratifying Resilience and Depression Longitudinally (STRADL)
Progress in understanding the causes of major depressive disorder has been slow. Dividing depression into subtypes, a process called stratification, could ultimately lead to faster progress.
We will stratify or divide individuals with MDD and depressive syndromes into more similar groups of people in UK Biobank.
Our aims are to:
1. Identify and describe specific subtypes of depression
2. Identify the causes underlying different types of depression using GWAS and MRI
3. Test whether resistance to depression (i.e. resilience) to depression can be accurately measured.
4. Identify the mechanisms underlying resilience using genetic and brain imaging data. This research seeks to use the medical, cognitive, imaging and genetic data from UKBiobank to study the mechanisms of common medical conditions and use them as a platform to better diagnosis. These aims are consistent with UK Biobank's. Providing this information will help to identify new drug targets for depression. Stratifying depression into more homogenous categories will provide better 'disease' targets for other research studies because there will be less lumping together of individuals with different causes for their illness within the same broad category of depression. We will test whether these sub-classes of depression and depressive symptom have neurobiological associations in UKbiobank by comparing them with depressed individuals as w whole, as well as controls, using MRI and genetic data.
We will firstly examine the associations of depression with cognition (baseline measures and web-based measures of attention and memory, for example), brain structure, function and connection strength (MRI).
We will examine the association of different depression types with biological intermediates (measurable variables important in the causation of depression) using a technique called polygenic profiling.
We will also compare resilient and non-resilient individuals. We are interested in the full UKbiobank cohort for most analyses - and the subgroup of UKbiobank with genetic and imaging (brain MRI) data for more detailed analysis.
We appreciate the time scale for the availability of genotyping and imaging data.
|Lead investigator:||Professor Andrew McIntosh|
|Lead institution:||University of Edinburgh|
6 related Returns
|Return ID||App ID||Description||Archive Date|
|1641||4844||A Genome-Wide Association Study Finds Genetic Associations with Broadly-Defined Headache in UK Biobank (N = 223,773)||19 Jul 2019|
|1664||4844||Genetic and Environmental Risk for Chronic Pain and the Contribution of RiskVariants for Major Depressive Disorder: A Family-Based Mixed-Model Analysis||23 Jul 2019|
|3050||4844||Genome-Wide Association Study Meta-Analysis of the Alcohol Use Disorders Identification Test (AUDIT) in Two Population-Based Cohorts||9 Dec 2020|
|749||4844||Genome-wide association study of alcohol consumption and genetic overlap with other health-related traits in UK Biobank (N=122117)||17 Oct 2017|
|1878||4844||Genome-wide association study of depression phenotypes in UK Biobank identifies variants in excitatory synaptic pathways.||16 Jan 2020|
|3049||4844||Genome-wide meta-analysis of depression in 807,553 individuals identifies 102 independent variants with replication in a further 1,507,153 individuals||9 Dec 2020|
|2922||Automated classification of depression from structural brain measures across two independent community-based cohorts||Stolicyn, A. et al||2020||Human Brain Mapping. 2020|