Brain age, epigenetic age, and neuropsychiatric disorders

Ageing is among the strongest risk factors for many neuropsychiatric diseases. However, ageing is not
uniform across the body. Our organs (e.g., brain), our bodily systems (e.g., cardiovascular, metabolic),
and even our DNA ages at different rates according to internal and external factors. This variability in biological ageing needs to be understood to make sense of how ageing influences, or is influenced by, neuropsychiatric disease. Importantly, some of the variability in biological ageing can be explained by our genetics.  

Artificial intelligence and machine learning models can be used to derive the biological age of the brain from MRI scans. The difference between biological age and chronological age is the ‘age gap’, and it represents deviation from the normative trajectory of ageing. Similarly, epigenetic age refers to an estimate of biological age based on patterns of DNA methylation, which is a chemical modification of DNA that influences gene expression without changing the underlying genetic sequence. Unlike chronological age, epigenetic age reflects how your body is
aging at a molecular level. Such biologically-informed ages can be used as a measure of internal system integrity, and a marker of dysfunction. 

Physical and mental health are fundamentally linked: As one is affected, the other is too. For example, accelerated biological ageing in later life is associated with greater risk of depression and anxiety. Ideally, epigenetic clocks and neuroimaging-derived brain age gaps could be harnessed to monitor and stratify risk of neurological and neuropsychiatric disease. However, because they each measure ageing very differently, there is conflicting evidence that brain age and epigenetic age are phenotypically associated. Nevertheless, both older-than-expected brain ages, and faster paces of epigenetic ageing, have been related to psychiatric conditions (e.g., Schizophrenia), mood disorders (e.g., depression) and neurodegenerative disorders (e.g., Alzheimer’s disease). Despite the (mixed) evidence of phenotypic relationships between brain age, epigenetic age, and neuropsychiatric disorders, their shared genetic bases remain incompletely understood. 

This project will broadly examine the shared and unique genetic architectures between brain and epigenetic ageing measures and their relationships with neuropsychiatric disorders. Genetic analyses can be harnessed to seek evidence regarding causality and biological mechanisms, which may also differ according to brain or epigenetic age estimation methods. Variations in the inputs used to calculate each biological age (i.e., different brain age measures, different epigenetic clocks) produce divergence in the estimation of the biological ages themselves. This has downstream effects on the attribution of genetic variants as contributing factors to neurological and epigenetic ageing, and how they relate to genetic variants determining neuropsychiatric disease. This divergence must be understood before we can disentangle causality between biological ageing and neuropsychiatric disease.

Some example research questions that can be explored in this project are:

• What are the genetic similarities between neuroimaging-derived brain ageing and epigenetic ageing?

• How do the genetic architectures of brain ageing and epigenetic ageing relate to external traits of interest, including (but not limited to): psychiatric traits (schizophrenia, autism spectrum disorder, attention-deficit hyperactivity disorder, obsessive-compulsive disorder), neurodegenerative traits (Alzheimer’s disease, Parkinson’s disease, fronto-temporal dementia), and ageing traits (frailty, longevity)?

• What are the conditionally independent genetic associations between each external trait and a network representing epigenetic and brain ageing?

• How do polygenic scores for brain ageing and each epigenetic clock relate to neuropsychiatric factors?

• Are there shared phenome-wide associations between brain ageing and epigenetic ageing? If so, do these point to shared biological pathways determining brain age and epigenetic age?

•  How much of ‘brain age’ and ‘epigenetic age’ measure system-wide integrity and how much is brain specific?

This project involves analysing clinical, genetic, and neuroimaging data (mostly tabular, with the potential for pre-processing of raw data if needed). Students will have the opportunity to perform genetic analyses including (but not limited to) genomic structural equation modelling, linkage disequilibrium score regression, and genetic network analysis. If required, students also can learn how to implement machine learning algorithms to estimate brain ages from large-scale neuroimaging datasets. Skills will be developed in the use of language-based environments for statistical computing (e.g., RStudio and bash).

This project offers the development of hard and soft skills necessary for a career in psychological, biomedical, or healthcare research. Students will gain invaluable experience in working among a team of peers while taking charge of their own independent project. There is strong potential for publication.