Diagnostic indicators and co-experience of mental health, neurodivergence, and learning difficulties for multi-omics discovery and prediction

Background

Psychiatric disorders (depression, anxiety, schizophrenia, eating disorders, etc.), neurodevelopmental conditions (autism, attention deficit/hyperactivity disorder), and specific learning difficulties (e.g., dyslexia, dyscalculia) frequently co-occur and represent a global burden to wellbeing. These conditions emerge during critical developmental windows, impacting a wide range of life outcomes. Large scale genetic studies have picked up shared signals between these conditions/neurodivergence, but interpreting this convergence is complicated by diagnostic overlap, heterogeneity in diagnostic presentation, and variation in data quality from multiple sources including questionnaires and health records. Psychological and neurocognitive functioning may also influence how these conditions are indicated in large datasets, such as factors that shape survey response patterns, access to healthcare, or the ability to take part in research. The co-experience of multiple conditions together (e.g., low mood in the context of autism or dyslexia) also needs to be considered to better guide research questions. This has implications for who is classified as being affected by a disorder in large-scale discovery research and for the equitable use of biomarker-based prediction.  

Aims

This project will take combined psychometric and genetic approaches to understand the diagnostic overlap and co-experience of psychiatric, neurodevelopmental, and learning conditions. The student will utilise large health datasets (UK Biobank, Generation Scotland, Swedish population registers, Psychiatric Genomics Consortium). The student will

1. Apply psychometric models to survey responses related to mental health, eating patterns, social interactions, and sensory perceptions. Develop explanatory models of diagnostic and symptom scales and relationships to demographic, cognitive, health, and multi-omics variables.
2. Conduct genetic association studies of diagnostic indicators from questionnaires, health records, and self-reports across datasets.  Conduct multivariate genomic analyses to assess contributions from core phenotypes versus ascertainment/methodological factors.
3. Integrate methylomic, proteomic and metabolomic data to identify intermediate biological layers between genes, environment, and phenotypes. Critically evaluate how precision medicine is shaped by participation and healthcare equity and role of misdiagnosis and underdiagnosis. Validate findings in replication datasets.
4. Work with people with lived experience to understand perspectives on the experience of individual diagnostic indicators and their co-occurrence with other conditions.

Training outcomes
Students are expected to have a BSc or MSc in a relevant field and an understanding of genetics and statistics. The student should be interested in health data science and the genetic epidemiology of health conditions. Training will be provided on:

1. Bioinformatics, multi-omics, epidemiology, and advanced statistical analyses of large-scale datasets.
2. Working in scientific programming languages, including using and writing analytic pipelines. 
3. Reproducible, scalable, and secure scientific computing
4. Critical evaluation of information, writing papers, and presenting work to academic audiences. 
5. Community engagement and participatory research methods.

References

• Cai, N. et al. Assessment and ascertainment in psychiatric molecular genetics: challenges and opportunities for cross-disorder research. Mol. Psychiatry 1–12 (2024) doi:10.1038/s41380-024-02878-x. 
• Grotzinger, A. D. et al. Mapping the genetic landscape across 14 psychiatric disorders. Nature 1–15 (2025) doi:10.1038/s41586-025-09820-3.
• Peyrot, W. J. et al. Distinguishing different psychiatric disorders using DDx-PRS. Nat. Genet. 1–10 (2026) doi:10.1038/s41588-026-02684-x. 
• Tiego, J. et al. Precision behavioral phenotyping as a strategy for uncovering the biological correlates of psychopathology. Nat. Ment. Heal. 1, 304–315 (2023). doi:10.1038/s44220-023-00057-5