Jaanika Kronberg

Interpreting the association of genetic variants with complex traits can be improved by gaining a greater understanding of the molecular consequences of these variants. Although genome-wide association studies (GWAS) for complex diseases routinely profile over one million individuals ^1–5 , studies of molecular traits have lagged behind. Here we performed a GWAS meta-analysis for 249 circulating metabolic traits in the Estonian Biobank and the UK Biobank in up to 619,372 individuals. We identified 88,127 common and low-frequency locus–trait associations from 8,398 loci that converged on shared genes and pathways. Using statistical fine mapping, systematic phenome-wide colocalization and cis -Mendelian randomization, we explored putative causal links between metabolic traits and disease outcomes. We predict that although plasma branched-chain amino acids (BCAAs) have been associated with type 2 diabetes in observational studies ^6,7 , lowering BCAA levels by targeting the BCAA catabolism pathway is unlikely to reduce type 2 diabetes risk. Leveraging our large sample size and high-quality genotype imputation, we found that 19.4% of the confidently fine-mapped variants had minor allele frequencies between 0.1 and 1%, and these variants were twofold enriched for predicted missense and splice-altering variants. Our results highlight the value of integrating low-frequency variants into genetic association studies.

Objective: To investigate associations between long-term environmental exposures, both external (ambient air pollution and built environment) and internal (circulating anthropogenic chemicals), and the human plasma metabolome, with the aim of generating biologically plausible hypotheses about affected metabolic pathways. Methods: We analyzed plasma from 989 Estonian Biobank participants using untargeted LC-HRMS (Metabolon HD4). External exposures ( PM _2.5 , PM ₁₀ , NO ₂ , ozone and built-environment metrics) were assigned using spatiotemporally resolved models developed in the EXPANSE project. Internal exposures were defined as ubiquitous anthropogenic compounds detected in the same metabolomics dataset. Associations between exposures and individual metabolites were quantified using left-censored regression models and then mapped to metabolite classes (Metabolon) and KEGG pathways. For enrichment analyses, one-sided Kolmogorov-Smirnov tests were applied to external exposures and Fisher’s exact tests to internal exposures. False discovery rate was controlled at 1% per exposure and database. Results: External air pollutants exhibited distinct metabolic patterns: Higher NO ₂ exposure was associated with enrichment of metabolites involved in tyrosine metabolism; higher ozone with monohydroxy and dicarboxylate fatty acids (consistent with lipid peroxidation); and higher PM _2.5 with acyl-carnitine subclasses and carbohydrate metabolism (glycolysis / gluconeogenesis / pyruvate). Built-environment associations were heterogeneous across metabolites and pathways. Internal anthropogenic chemicals showed broader metabolic associations than external exposures, involving a larger number of metabolites and metabolic classes. PFAS (PFOA, PFOS) were associated with long-chain polyunsaturated fatty acids (n3/n6) and lysophospho-lipids. Associations with 4-hydroxychlorothalonil, a fungicide, pointed to androgenic steroid metabolites and alpha-linolenic acid metabolism. The phenolic 2,4-di-tert-butylphenol, a plastic associated chemical, showed widespread associations with lipid classes, suggesting disruption of membrane remodeling and fatty acid handling. Conclusion: Long-term environmental exposures, both external and internal, are measurably reflected in the human plasma metabolome. Across exposure domains, recurrent signals involved lipid metabolism, membrane composition, and oxidative stress–related pathways, highlighting these as common biological targets of environmental exposures. The findings generate testable hypotheses, including nitrosative stress–related alterations for NO ₂ , lipid peroxidation for ozone, energy-metabolism perturbations for PM _2.5 , potential endocrine activity for chlorothalonil metabolites, and possible obesogenic effects of 2,4-di-tert-butylphenol.

A high-throughput platform of nuclear magnetic resonance (NMR) spectroscopy blood metabolite data from 30-90 years old participants of Estonian Biobank ( N = 150, 023) is used to develop and validate a biomarker score for mortality and the corresponding biological age estimate. We define biological age as the age where an individual’s survival probability given their covariate profile matches the survival probability of an average individual in the population. We estimate the survival probabilities parametrically based on Gompertz model using the newly developed metabolite biomarker score and common risk factors. The resulting biological age estimate, SurvMetaboAge , is a powerful predictor of both overall and cause-specific mortality. One year of biological age acceleration (BAA, difference between individual’s biological and chronological age) is associated with 17% (95% CI 15%–18%) and 12% (95% CI 11%–13%) increase of hazards for overall mortality in the validation set for men and women, respectively. An appealing interpretation of BAA can be provided, demonstrating that it is an easily interpretable predictor of mortality encompassing metabolite profile and common risk factors in a single measure with a potential in risk stratification and risk communication.

Parent-of-origin effects (POEs) occur when the effect of a genetic variant depends on its parental origin¹. Traditionally linked to genomic imprinting, POEs are believed to occur due to parental conflict over resource allocation to offspring, resulting in opposing parental influences². Despite their importance, POEs remain underexplored in complex traits, owing to the lack of parental genomes. Here we present an approach to infer the parent of origin of alleles without parental genomes, leveraging interchromosomal phasing, mitochondrial and X chromosome data, and sex-specific crossover in siblings. Applied to the UK Biobank, this enabled parent-of-origin inference for up to 109,385 individuals. Genome-wide association study scans for 59 complex traits and over 14,000 protein quantitative trait loci contrasting maternal and paternal effects identified over 30 POEs and confirmed more than 50% of known associations. More than one third of these showed opposite parental influences, especially for traits related to growth (for example, IGF1 and height) and metabolism (for example, type 2 diabetes and triglyceride levels). Replication in up to 85,050 individuals from the Estonian Biobank and 42,346 offspring from the Norwegian Mother, Father and Child Cohort Study (MoBa) validated 87% of testable associations. Overall, our findings highlight the contribution of POEs to complex traits and support the parental conflict hypothesis, providing compelling evidence for this understudied evolutionary phenomenon.

Background/Objectives: Metabolomics, in combination with genetic data, is a powerful approach to study the biochemical consequences of genetic variation. We assessed the impact of human gene knockouts (KOs) on the metabolite levels of Estonia Biobank (EstBB) participants and integrated the results with electronic health record data. Methods: In 150,000 EstBB genotyped participants, we identified 723 KOs with 152 different predicted loss of function (pLoF) variants in 115 genes. For those KOs and 258 controls, 1387 metabolites were profiled using ultra-high-performance liquid chromatography–tandem mass spectrometry. Results: We identified 48 associations linking rare pLoF variants in 22 genes to 43 metabolites. Out of 48 associations, 27 (56%) were found in genes that cause inborn errors of metabolism. The top associations identified in our analysis included genes and metabolites involved in the degradation pathway of the pyrimidine bases uracil and thymine (DPYD and UPB1). We found DPYD gene KOs to be associated with elevated levels of Uracil, confirming that DPD-deficiency is a leading cause of severe 5-Fluorouracil toxicity. Overall, 54% of reported associations are gene targets of approved drugs or bioactive drug-like compounds. Conclusions: Our findings contribute to assessing the impact of human KOs on metabolite levels and offer insights into gene functions, disease mechanism, and drug target validation.

Large biobanks have set a new standard for research and innovation in human genomics and implementation of personalized medicine. The Estonian Biobank was founded a quarter of a century ago, and its biological specimens, clinical, health, omics, and lifestyle data have been included in over 800 publications to date. What makes the biobank unique internationally is its translational focus, with active efforts to conduct clinical studies based on genetic findings, and to explore the effects of return of results on participants. In this review, we provide an overview of the Estonian Biobank, highlight its strengths for studying the effects of genetic variation and quantitative phenotypes on health-related traits, development of methods and frameworks for bringing genomics into the clinic, and its role as a driving force for implementing personalized medicine on a national level and beyond.

Nationwide health records linked to population biobanks can expand genetic discovery into clinical phenotypes that are poorly captured in hospital-centred datasets. We performed genome-wide association analyses of 5,491 ICD-10-based disease phenotypes in 206,159 Estonian Biobank participants using imputed genotype data across 18.8 million single-nucleotide and insertion-deletion variants. Across the disease phenome, we identified 3,222 genome-wide significant loci, with strongest added value for outpatient-enriched, recurrent, and earlier-onset conditions. Fine-mapping prioritised candidate causal variants across study-wide significant loci, while coding variant analyses identified 754 protein-altering variant-trait associations outside the HLA region, including high-confidence signals in various dermatological, anaemia, congenital, and metabolic traits. A separate HLA analysis identified 744 HLA-trait associations across infectious, autoimmune, and skin-related phenotypes. As an example of discovery in an understudied phenotype, we highlight pityriasis versicolor, a superficial fungal infection with 34 loci in EstBB-FinnGen meta-analysis, including a rare Estonian-enriched splice-disrupting TNFSF15 variant. These results establish EstBB as a resource for mapping genetic architecture across clinically diverse and underrepresented disease phenotypes.

The purpose of this study was to find metabolic changes associated with incident hypertension in the volunteer-based Estonian Biobank. We used a subcohort of the Estonian Biobank where metabolite levels had been measured by mass-spectrometry (LC-MS, Metabolon platform). We divided annotated metabolites of 989 individuals into KEGG pathways, followed by principal component analysis of metabolites in each pathway, resulting in a dataset of 91 pathway components. Next, we defined incident hypertension cases and controls based on electronic health records, resulting in a dataset of 101 incident hypertension cases and 450 controls. We used Cox proportional hazards models and replicated the results in a separate cohort of the Estonian Biobank, assayed with LC-MS dataset of the Broad platform and including 582 individuals. Our results show that body mass index and a component of the carbon metabolism KEGG pathway are associated with incident hypertension in both discovery and replication cohorts. We demonstrate that a high-dimensional dataset can be meaningfully reduced into informative pathway components that can subsequently be analysed in an interpretable way, and replicated in a metabolomics dataset from a different platform.