What we are doing

​​​​​​

​​

1. Systematic genetic screening of metabolic proteins and metabolites

​

We deploy large-scale Mendelian randomisation (MR) to test whether genetically predicted variation in circulating metabolic proteins or metabolites influences risk of SMI. We integrate:

• Proteomic QTL (pQTL) and metabolite QTL (metQTL) datasets

• Genome-wide association studies (GWAS) of SMI

• Large population cohorts including UK Biobank, ALSPAC and Generation Scotland

​

This work is built upon our previous study interrogating causal relationship between metabolite abundance and Major Depression: see here.

​

2. Multi-omic target validation or molecular targets that pass initial MR screening, we apply:

• Multi-trait fine-mapping

• Genetic co-localisation (eQTL/pQTL/chromatine accessibility/methQTL with SMI loci)

• Network and pathway enrichment analyses

• Drug target mapping

​

This triangulation approach reduces false positives and increases confidence that prioritised proteins are biologically meaningful. This work has been published in Biological Psychiatry: see here.

​

3. Trans-diagnostic and phenome-wide analyses

We evaluate whether inflammatory targets act:

• Across diagnostic categories (trans-diagnostic effects)

• In specific subgroups

• In relation to physical–mental health comorbidity

​

Using MR-PheWAS approaches in large biobanks, we assess potential beneficial and adverse downstream consequences of modifying each target.

4. Diversity and global applicability

​

Where possible, we incorporate GWAS and molQTL data from diverse ancestries, including African cohorts, to improve generalisability and mechanistic resolution.​​​

What we are planning to do

​​​​​

​​

Over the lifetime of ImmunoMIND we will:

• Expand screening to thousands of immune-related proteins and metabolites as datasets grow

• Integrate longitudinal -omic data and linked electronic health records to understand drug mechanism

• Provide validated target lists to the Therapeutic Personalisation platform for drug matching and repurposing

• Integrate recent advances in cell-type reference data to interrogate -omic associations with SMI that are specific to biological context.​

​

Our ambition is to “funnel down” from hundreds of plausible inflammatory exposures to a small set (~5%) of genetically supported causal targets.

Find out more

​​​​​

​​​

Key methodological foundations include:

• Two-sample Mendelian randomisation

• Genetic co-localisation and fine-mapping

• Transcriptomic and protein–protein interaction network analysis

• MR-PheWAS frameworks

Relevant prior work from our teams has demonstrated the feasibility of large-scale immune protein screening for psychiatric outcomes.

• Nisbet, L., …, Wray, N. R., McIntosh, A. M., & Shen, X. (2025). Integrating multi-omic summary data identifies candidate molecular mechanisms for Major Depression. Biol Psychiatry. https://doi.org/10.1016/j.biopsych.2025.11.024

• Shen, X. (2025). Limitations and Potential of Polygenic Risk Scores for Major Psychiatric Disorders. Biol Psychiatry, 98(6), 444-445. https://doi.org/10.1016/j.biopsych.2025.07.007

• Shen, X., …, Wray N. R., McIntosh, A. M. (2025). A methylome-wide association study of major depression with out-of-sample case-control classification and trans-ancestry comparison. Nat Ment Health, 3(10), 1152-1167. https://doi.org/10.1038/s44220-025-00486-4

• Shen, X., et al. (2025). Association between polygenic risk for Major Depression and brain structure in a mega-analysis of 50,975 participants across 11 studies. Mol Psychiatry. https://doi.org/10.1038/s41380-025-03136-4

• Xu, E. Y., ..., McIntosh, A. M., Shen, X., & Whalley, H. C. (2025). Epigenetic and Structural Brain Aging and Their Associations With Major Depressive Disorder. Biol Psychiatry Glob Open Sci, 5(6), 100577. https://doi.org/10.1016/j.bpsgos.2025.100577

Get involved

​​​​​​

​​

There are no opportunities at the moment. Check again soon!

​

​

​

​