David Stacey

The cellular genetic architecture underlying major depressive disorder (MDD) remains poorly understood, partly because bulk brain transcriptomic analyses obscure cell-type-specific regulatory effects. By integrating single-nucleus brain eQTL data from SingleBrain with large-scale MDD genome-wide association data through transcriptome-wide Mendelian randomisation, we identified 246 MDD-linked gene expression signals (eGenes) across seven major brain cell types, with prominent signals in excitatory neurons, inhibitory neurons, and astrocytes. Colocalisation prioritised 89 robust eGenes across 66 independent loci, with a risk-increasing effect of higher endothelial BDNF expression on MDD and marked heterogeneity across cellular subtypes. Age-at-onset analyses uncovered distinct genetic signatures, implicating ZCCHC7 and RASGRP1 in early-onset MDD and RAB27B and ITIH4 in late-onset MDD. Prioritised genes were enriched for evolutionary constraint and Mendelian disease associations, converged on synaptic, neurodevelopmental, and metabolic processes, and included seven clinically supported Tier 1 drug targets. These findings provide a high-resolution cellular map of risk genes of MDD and nominate candidate therapeutic targets.

Background: The Five Factor Model (FFM) personality meta-traits, stability and plasticity, are thought to capture broad individual differences in the parameters of goal-directed, cybernetic mechanisms in the brain, which have implications for transdiagnostic psychopathology risk. It remains unknown whether these meta-traits represent genetically coherent dimensions. Here we conduct a genomic investigation into the FFM meta-traits and characterize their association with 11 psychiatric conditions. Methods: We combined recent genome-wide association study (GWAS) summary statistics with data from the 23andMe Research Institute (European ancestry; total N = 279,240–682,707) and applied Genomic Structural Equation Modelling to conduct a multivariate GWAS of the FFM meta-traits. We then conducted transcriptome-wide association studies (TWAS) and stratified linkage disequilibrium score regression (S-LDSC) to prioritize candidate effector genes and brain cell types for personality traits. Finally, we used LDSC regression and CAUSE Mendelian randomization to characterize the relationship between personality traits and 11 psychiatric conditions. Results: We identified 83 (14 novel) and 13 (4 novel) independent genome-wide significant loci for stability and plasticity respectively. TWAS prioritized hundreds of candidate effector genes for personality, and S-LDSC showed heritability enrichment across several neuronal cell types. Finally, genetic correlation and Mendelian randomization analyses provided genetic evidence suggesting bidirectional links between stability and psychopathology: higher genetic propensity for stability was protective against psychopathology, whereas greater genetic liability to psychopathology was associated with reduced stability. Conclusions: The personality meta-traits capture modest shared architecture across personality, with genetic liability for lower stability potentially representing a broad transdiagnostic link between personality and psychiatric vulnerability.

Inflammatory pathways are implicated in depression, but the specific immune proteins and causal variants involved remain unclear. We tested potential causal relationships between 91 immune-related plasma proteins and depression using Generalised Summary-data-based Mendelian Randomisation (GSMR). We identified an association between CD40 protein levels and depression (OR: 0.95, 95% CI: 0.94–0.97, p = 1.71 × 10 ⁻¹¹ ), explained largely by variants located near the CD40 gene. Follow-up colocalisation analyses indicated that the CD40 protein and depression associations in this region were attributable to different lead variants that are often inherited together, suggesting the GSMR association was confounded by correlated genetic signals. Analyses using gene expression data prioritised SLC12A5 , not CD40 , as the most likely gene driving the depression signal at this locus. SLC12A5 produces KCC2, a neuronal potassium-chloride transporter primarily expressed in the brain. Scanning a wide range of health traits showed the CD40 lead variant was predominantly associated with inflammatory disorders, while the depression lead variant was more strongly linked to psychiatric conditions. Overall, these results emphasise the value of combining Mendelian randomisation with colocalisation to separate shared biology from correlated signals in genomic regions where multiple traits map to the same locus (pleiotropy). In this case, plasma CD40 protein levels are unlikely to have a direct causal role in depression, whereas SLC12A5 -mediated effects may contribute to its underlying biology. Further functional and multi-omic studies are needed to clarify immune-brain interactions and identify therapeutic targets for depression.

1. Clozapine is the gold-standard medication for treatment-resistant schizophrenia (TRS), which comprises ∼30% of all people with schizophrenia (SCZ). However, its therapeutic potential is severely...

The Five-Factor model (FFM) personality traits (agreeableness, conscientiousness, extraversion, neuroticism and openness) capture stable individual differences in thinking, feeling and behaviour. It has been shown that the FFM traits share variance through two-higher order “meta-traits”, stability and plasticity. It remains unknown, however, whether these meta-traits can capture the shared genetic architecture of the FFM personality traits. Here we combined recent genome-wide association study (GWAS) summary statistics with data from the 23andMe Research Institute (European ancestry; total N = 279,240-682,707) and applied Genomic Structural Equation Modelling, identifying two latent genetic factors consistent with stability and plasticity. We then performed a multivariate GWAS on these factors to capture genetic loci with shared effects across the FFM traits, identifying 81 and 13 independent genome-wide significant loci for stability and plasticity respectively. Transcriptome-wide and cell-type enrichment analyses prioritised candidate effector genes and indicated broad brain involvement for personality traits. Finally, genetic correlation and pleiotropy-aware Mendelian randomisation analyses provided genetic evidence suggesting bi-directional links between stability and psychopathology: higher genetic propensity for stability was protective against psychopathology, whereas greater genetic liability to psychopathology was associated with reduced stability. These results enhance our understanding of the shared genetic architecture underlying personality traits and their overlap with psychopathology.

0. Clozapine is the gold-standard for treatment-resistant schizophrenia despite its severe metabolic complications, including metabolic syndrome (MetS) and type 2 diabetes (T2D) risk. A better understanding of the genetic factors influencing clozapine pharmacokinetics and their relationship to metabolic risk could help inform precision medicine approaches to clozapine prescribing. Using a series of genetic-epidemiological approaches, we aimed to identify candidate biomarkers associated with clozapine-induced metabolic dysfunction. We first used two-sample Mendelian randomization (MR) leveraging genome-wide association summary data to investigate evidence of causal relationships between clozapine metabolism and cardiometabolic traits. These analyses indicated that higher plasma clozapine levels and a higher clozapine-norclozapine ratio were both associated with a higher risk of T2D and higher blood pressure. We then applied a Phenome-scan-colocalization-MR pipeline to identify traits influenced by clozapine-metabolism loci that might serve as biomarkers of cardiometabolic risk. This pipeline identified 28 colocalizing candidate biomarkers associated with clozapine metabolising genetic loci. Subsequent MR highlighted associations for 16 of these 28 biomarker candidates with cardiometabolic outcomes, which included haematological markers and excretory traits (e.g. gamma-glutamyl transferase, red cell distribution width, and urea). These findings may inform the development of biomarker-guided monitoring approaches for risk stratification and early intervention, enabling a shift from reactive monitoring to predictive approaches in managing clozapine-induced metabolic dysfunction with appropriate clinical validation. These findings may also help to mitigate the risk of metabolic dysfunction associated with other antipsychotic medications.

Background: Interventions to prevent type 1 diabetes (T1D), an immune-mediated disease requiring lifelong treatment, remain limited. We sought to identify novel therapeutic targets for T1D through Mendelian randomization and colocalization using immune cell-derived instruments. Methods: We selected locally acting genetic variants from 14 transcriptomic studies to instrument expression of 8998 genes measured in immune cell types. Outcome associations were obtained from a T1D genome-wide association study that included 18,942 cases and 501,638 controls. Follow-up analyses included assessments of horizontal pleiotropy and novelty, as well as phenome-wide scans with colocalization (PheWAS-coloc) of instrumental variants. Results: We prioritized 21 genes ( CLNK , EED , LZTFL1 , MGAT4A , NAA38 , NFKB1 , PHACTR4 , PHLPP2 , PLEKHA1 , P2RY12 , REST , RGS14 , SERPINB6 , SESN3 , SLC25A29 , SPAG1 , STIM2 , THEMIS , TMEM80 , VSIR , ZNF217 ) that have not been identified in previous T1D genome-wide association studies. Notably, higher genetically predicted VSIR (encoding the immune checkpoint protein VISTA) was associated with decreased risk of T1D, providing human genetic support that complements existing animal-model evidence for VISTA’s protective role in autoimmune diseases. We also prioritized P2RY12 , which is targeted by several approved drugs, revealing a possible repurposing opportunity. PheWAS-coloc analyses further linked P2RY12 expression to Epstein-Barr virus EBNA-1 antibody levels, implicating a pathway relevant to autoimmunity. Conclusions: Our findings provide novel avenues for drug development and repurposing to prevent or delay T1D onset.

Background: Ovarian cancer (OC) is often diagnosed at an advanced stage when prognosis is poor. We aimed to identify blood plasma proteins predictive of OC risk. Methods: We conducted proteome-wide Mendelian randomisation (MR) analyses using summary-level protein quantitative trait locus data covering 2337 plasma proteins, and genome-wide association data on OC and its subtypes (up to 25,509 cases) from the Ovarian Cancer Association Consortium. Wald ratio or inverse-variance weighted MR analysis was used as the primary method, depending on the number of instruments. We evaluated pleiotropy using MR-Egger intercept test and leave-one-out analysis. Results: From 2337 plasma proteins, 12 were associated (p < 7.4 × 10⁻⁵) with OC or its subtypes. Robust evidence linked follitropin subunit beta (FSHB) with endometrioid OC (per SD higher, OR 2.41, 95% CI 1.56, 3.71). Associations for the other 11 proteins could be explained by pleiotropy from ABO or MAPT-AS1 loci. We identified 12 suggestive associations with OC or its subtypes at nominal threshold (p < 0.05), involving 11 plasma proteins, with no evidence of pleiotropy from leave-one-out and MR-Egger intercept tests (P _intercept > 0.17). Potential drug targets were identified for follitropin receptor and eight other proteins. Conclusion: Our study suggests FSHB and 11 additional plasma proteins as of potential interest in OC (or subtypes) prognosis, mostly representing potentially druggable targets.

Inflammatory pathways are implicated in depression, but the specific immune proteins and causal variants involved remain unclear. This study investigated potential causal relationships between 91 immune-related plasma proteins and depression using generalized summary Mendelian randomisation. We identified a robust association between CD40 protein levels and depression (OR: 0.95, 95% CI: 0.94 - 0.97, p = 1.71 × 10⁻¹¹), primarily driven by cis- acting variants. However, pairwise statistical colocalisation analyses of the CD40 locus indicated that CD40 protein and depression had distinct – though not independent – lead variants, suggesting the Mendelian randomisation signal was confounded by linkage disequilibrium. Analyses using expression quantitative trait locus data from the Genotype-Tissue Expression project prioritised SLC12A5 , not CD40 , as the most likely effector gene for depression risk at the locus. SLC12A5 encodes a potassium chloride co-transporter preferentially expressed in brain tissue, consistent with a role in depression. A phenome-wide association study showed the CD40 protein lead variant was primarily associated with inflammatory disorders, while the depression lead variant was more strongly linked to psychiatric conditions. Our results emphasise the importance of combining Mendelian randomisation with colocalisation analyses to disentangle pleiotropic effects at loci with complex genetic architecture, such as CD40/SLC12A5 . While plasma CD40 protein levels are unlikely to play a causal role in depression, SLC12A5 -mediated effects may contribute to its pathophysiology. These findings highlight the need for further functional and multi-omic studies to clarify immune-brain interactions and identify therapeutic targets for depression.