Why do some people develop psychiatric disorders while others do not? Despite decades of research, this question remains difficult to answer. Psychiatric disorders are shaped by multiple, interacting influences, including genetics and environmental factors. Untangling how such risk factors work together remains a central challenge for the field (Burmeister et al. 2008), yet doing so could help improve diagnosis, treatment, and prevention.
Genome-wide association studies (GWAS) have identified many genetic variants linked with mental health, but these only account for a small fraction of heritability (Trubetskoy V et al. 2022; Demontis D et al. 2023; Donnelly N and Foley E, 2025). Mendelian randomization (MR) is a genetic epidemiological method that uses GWAS summary data to assess whether one factor might directly influence another (Emdin CA et al. 2017; Crick D, 2023). Identifying risk factors that likely cause a disorder opens up the opportunity for the development of new, targeted treatments and/or prevention tactics.
Despite its promise as a technique, a comprehensive database detailing MR evidence for psychiatric disorders is currently lacking. To overcome this, Li et al. (2025) have developed a new comprehensive database for researchers called PsyRiskMR, designed to facilitate the analysis of risk factors for psychiatric disorders.
Understanding what drives mental health disorders is complex. PsyRiskMR is a new database designed to help researchers uncover potential risk factors and causal links.
Methods
The authors used publicly available GWAS summary data from the Psychiatric Genomics Consortium to study the 10 most common psychiatric disorders: attention deficit disorder (ADHD), Alzheimer’s disease, anxiety disorder, bipolar disorder, eating disorders, depression, obsessive-compulsive disorder (OCD), post-traumatic stress disorder (PTSD), and schizophrenia.
They searched several sources for risk factors, categorised by risk factor type:
- Risk phenotype = Traits or characteristics (like personality or lifestyle factors) that might influence the risk of psychiatric disorders.
- Risk brain imaging = Measures from brain scans that could indicate structural or functional differences linked to mental health conditions.
- Bulk-tissue xQTL = Genetic variants in tissue that may affect gene activity and be linked to psychiatric disorders.
- Cell-specific xQTL = Genetic variants that affect specific types of cells (neurons, microglia, stem cells, and lymphocytes), helping identify which cells contribute to mental health risks.
MR analyses were then carried out to investigate whether these risk factors might causally influence the 10 psychiatric disorders. The analyses included statistical corrections to reduce false positives and additional sensitivity checks to confirm the results.
Results
PsyRiskMR provides a useful interface for researchers to examine MR results for psychiatric disorders. It consists of four modules and the authors plan to update the data on the website every 6 months.
Seventy-one psychiatric disorder traits were selected, along with 3,935 brain imaging measures and more than 30 genetic datasets from brain tissue and specific cell types. These covered six different xQTL types.
Risk phenotypes & psychiatric disorders
Using MR, the authors found 16 risk traits with strong links to psychiatric disorders. Many of the traits were associated with more than one disorder. For example, extraversion, educational attainment, and neuroticism were associated with both anxiety and bipolar disorder. This demonstrates the complexity of the association between mental health risk factors.
Risk brain imaging & psychiatric disorders
Seven brain imaging traits were associated with psychiatric disorders. Interestingly, there was an overlapping MR result between schizophrenia and PTSD (i.e., resting state magnetic functional imaging connectivity), suggesting that this part of the brain is involved in both disorders.
Bulk-tissue xQTL & psychiatric disorders
There was strong evidence of a causal link between 269 risk genes and 5 disorders (ADHD, depression, Alzheimer’s disease, bipolar disorder, schizophrenia). Twenty-five of these genes were associated with more than one disorder.
Cell-specific xQTL & psychiatric disorders
Eighty-four genes were causally associated with psychiatric disorders. However, only 45 of these genes showed significant overlap with those found in bulk tissue. This shows the added value of looking at specific cell types.
PsyRiskMR example: Schizophrenia
On the PsyRiskMR website, specific disorders of interest can be selected. If, for example, one selects schizophrenia, you will see that several phenotypic risk factors have been identified (i.e., trauma exposure, type 1 diabetes, neuroticism, smoking, being unable to work because of disability, brain imaging resting-state functional magnetic resonance imaging connectivity and cortical thickness).
PsyRiskMR allows users to explore the many factors that may contribute to psychiatric disorders, from genetics and brain structure to lifestyle and environment.
Conclusions
The creation of PsyRiskMR has provided an essential tool for researchers who work on investigating the complex and multifactorial risk factors for the 10 most common mental disorders. The authors say:
We hope that PsyRiskMR will become a user-friendly platform facilitating research into the underlying mechanisms of psychiatric disorders and offering valuable insights for their improved diagnosis, prevention and treatment.
PsyRiskMR opens the door for researchers to better understand mental health, helping turn complex data into actionable insights for diagnosis, treatment, and prevention.
Strengths and limitations
A key strength of this study is its creation of a web portal that brings together genetic data from multiple sources for all the main mental health risk factors categories. This makes PsyRiskMR an extremely valuable resource and may help guide future prevention and treatment efforts.
The authors also compared the genes identified for schizophrenia in PsyRiskMR with two other similar resources. Surprisingly, 63 of these genes were unique to PsyRiskMR. However, the authors made no attempt to explain the low level of overlap between their resource and other recently developed resources in their paper.
Other limitations include the focus on genetic studies from people of European ancestry (an unfortunately very common limitation in genetic epidemiology research). While this is a necessary evil based on currently available data and is currently required to ensure maximisation of sample size and MR validity, it does mean that their findings cannot be generalised to other ethnic groups. This is particularly relevant for schizophrenia, as some non-white ethnicities carry different risk levels and factors (Kirkbride et al 2017).
Some datasets in PsyRiskMR have quite small sample sizes. Therefore, many of the MR analyses were underpowered. This was particularly true of the trans-xQTL data and is an important issue which can reduce the reliability of the casual analyses.
PsyRiskMR offers a powerful research resource, but its coverage and generalisability have limits that users should consider.
Implications for practice
This study is far from influencing clinical practice. While it achieved its main purpose of providing a resource for mental health risk factor research, it will be some time before findings from studies using PsyRiskMR inform clinical care.
In the future, if researchers using PsyRiskMR can provide strong enough evidence that certain risk factors directly cause/contribute to psychiatric disorders, this could lead to new treatment approaches and prevention efforts. For example, identifying modifiable lifestyle factors or biomarkers could help guide early interventions or personalised care.
From a research perspective, PsyRiskMR is a particularly valuable tool. As psychiatric epidemiologists, we are particularly interested in this study because having all relevant data on risk factors and outcomes in one accessible place can speed up research and reduce duplication. It can also serve as an educational resource for researchers, clinicians, and others seeking to understand the genetic and environmental contributions to psychiatric disorders.
The database will continue to evolve as new data become available, helping maintain its relevance and usefulness for future studies. Over time, it may help bridge the gap between research and clinical practice, but careful validation is needed before any findings are applied in healthcare settings.
This database supports research into mental health risk factors while highlighting that clinical applications remain a future goal.
Statement of interests
Sarah wrote the first draft of this blog and has no competing interests to declare. Eimear is a coordinator for the Mental Elf and worked on the second draft on the blog. She has no conflicts of interest to declare.
Editor
Edited by Éimear Foley. AI tools assisted with language refinement and formatting during the editorial phase.
Links
Primary paper
Li X, Shen A, Fan L, Zhao Y, Xia J (2025) PsyRiskMR: A comprehensive resource for identifying psychiatric disorder risk factors through Mendelian Randomisation. Biological Psychiatry 98: 126-134. DOI: 10.1016/j.biopsych.2024.11.018
Other references
Burmeister M, McInnis MG, Zollner S (2008) Psychiatric genetics: progress amid controversy. Nat Rev Gen 9:527-540. DOI: 10.1038/nrg2381
Trubetskoy V, Pardinas AF, Ting Q et al (2022) Mapping genomic loci implicates genes and synaptic biology in schizophrenia. 604: 502-508. DOI: 10.1038/s41586-022-04434-5
Demontis D, Bragi Walters G, Athanasiadis G et al (2023) Genome-wide analyses of ADHD identify 27 risk loci, refine the genetic architecture and implicate several cognitive domains. Nat Gens 55:198-208. DOI: 10.1038/s41588-022-01285-8
Emdin CA, Khera AV, Kathiresan S (2017) Mendelian Randomization. JAMA Guide to Statistics and Methods 318(19). doi:10.1001/jama.2017.17219
Donnelly, N and Foley, E. Do psychiatric disorder genes overlap with their drug targets? And does this matter? The Mental Elf, 27 August 2025
Crick, D. Does what you eat affect how you feel? The Mental Elf, 08 June 2023
Kirkbride J B, Hameed, Y, Ioannidis K et al (2017) “Ethnic minority status, age at immigration and psychosis risk in rural environments: evidence from the SEPEA study. Sz Bull 43(6) 1251-1261. DOI: 10.1093/schbul/sbx010
