Biology subjects

Kutalik, Z.

Publications and source records attributed to Kutalik, Z..

At least 19 recordsLinked to original sources

bioRxiv2026

Triangulating evidence to detect signatures of stabilizing selection acting on protein levels in humans

Stabilizing selection is thought to maintain quantitative traits near evolutionary optima, yet the extent to which it constrains human molecular phenotypes remains largely unknown. Here, we investigate signatures of stabilizing selection on the circulating levels of 2'923 plasma proteins in up to 50'000 UK Biobank participants by integrating complementary evidence across evolutionary timescales. We combine genetic signatures of selection accumulated over generations with associations between protein abundance and contemporary fitness-related outcomes. Genetic evidence was assessed using the frequency-dependent architecture of common variants with LDpred2 and a newly developed likelihood-based framework leveraging rare loss-of-function variant burdens. Contemporary evidence was obtained through quadratic associations between protein abundance and three fitness proxies: reproductive success, somatic health, and parental longevity. Across all approaches, negative selection coefficients predominated. Among significant estimates, 100% of common-variant and 92.4% of rare-variant estimates were consistent with stabilizing selection, compared with 99.8% for somatic health, 97.7% for parental longevity, and 89.3% for reproductive success. Despite capturing distinct aspects of selection, the approaches showed partial convergence. Rare-variant-based estimates correlated with health-score-derived coefficients (rho = 0.36, p = 5.2 x 10^-6) and independent measures of evolutionary constraint, including protein heritability, loss-of-function intolerance, and cross-species transcript evolutionary rates. Together, our findings suggest that stabilizing selection is widespread across the human plasma proteome and highlight the value of integrating genetic architecture with fitness-related phenotypes to understand evolutionary constraints on molecular variation.

evolutionary biology↗

bioRxiv2026

Evaluating anonymized genome re-identification using polygenic predictions and its implications for data privacy

Re-identification by phenotypic prediction aims to determine whether a genome belongs to a specific individual by comparing the individuals known traits with those predicted from the genome. This type of tracing attack is widely discussed in the genomic privacy literature, yet previous studies have been criticized for overstating its practical risks. Over the past decade, genome-wide association studies (GWAS) with increasing sample size improved the accuracy of phenotypic prediction, potentially enhancing such attacks. To quantify their real-world threat, we developed a probabilistic framework that estimates the likelihood of a match between an individuals observed traits and polygenic scores (PGS) derived from a genome, while accounting for prediction accuracy and genetic and environmental correlations between the traits. We benchmarked this re-identification method and examined how the prior probability (reflecting the a priori chance that a random genome and set of traits correspond to the same person) affects performance. Finally, we assessed whether sensitive information could be inferred through this attack by attempting to predict multiple sensitive haplotypes, such as APOE-{varepsilon}4 (linked with Alzheimers disease). Our re-identification method outperformed a state-of-the-art tool, and reached a precision above 99% for a recall of 40% when considering a prior of 50%. However, after considering real-world settings, we estimated that realistic priors would not exceed 4 x 10-4%, resulting in a precision lower than 0.13% at the same recall (40%). The inference of sensitive genotypes also proved ineffective, as achieving a precision above 50% for identifying APOE-{varepsilon}4 carriers was only possible at a recall below 20%. To conclude, although re-identification by phenotypic prediction is technically feasible, our findings indicate that its effectiveness in real-world conditions is limited. These results counterpoint to earlier claims of severe genomic privacy risks and offer guidance for policymakers, biobank administrators, and research participants.

genetics↗

bioRxiv2026

CYP4F2-mediated ω-hydroxylation of 1-deoxysphingolipids reveals a new hepatic detoxification pathway

1-deoxysphingolipids (1-deoxySLs) are atypical, cytotoxic sphingolipids (SL) formed by the serine palmitoyltransferase through the alternative use of L-Alanine over its canonical substrate L-Serine. Elevated plasma levels of 1-deoxySLs have been implicated in metabolic and neurodegenerative diseases. Due to the missing C1 hydroxyl group, 1-deoxySLs cannot be converted into complex sphingolipids nor degraded via the canonical SL catabolic pathways. However, previous reports suggested a cytochrome P450 mediated {omega}-hydroxylation of 1-deoxySLs as a potential detoxification mechanism although the exacts downstream metabolism of these lipids remained unclear. We combined genome-wide association analysis with targeted lipid analysis to identify genes involved in 1-deoxySL metabolism. Functional validation was performed in cell culture models, enzyme assays, and through quantitative high-resolution mass spectrometry using isotope labelled synthetic standards.We identified a strong association between the CYP4F2 rs2108622 variant and plasma 1-deoxySL, implicating CYP4F2 is involved in 1-deoxySL metabolism. We demonstrated that CYP4F2 catalyzes the {omega}-hydroxylation of 1-deoxysphinganine, forming a previously uncharacterized hydroxylated sphingoid base. In liver cells, this metabolite was further metabolized via three distinct pathways: one forming the N-acyl, a second involving omega acylation and third resulting in omega carboxylation. All reactions generated a new spectrum of 1-deoxysphingolipids that are based on {omega}-hydroxylated 1-deoxySA as a precursor. The metabolic steps were confirmed by structural validation using synthetically prepared external standards. Importantly, {omega}-hydroxylation significantly attenuated the acute cytotoxicity of 1-deoxySLs in liver cells, indicating that this modification is the initiating step of a multi-branched metabolic clearance pathway. This study identifies CYP4F2 as a key enzyme initiating the hepatic clearance of atypical 1-deoxySLs, mitigating their cellular toxicity and revealing multiple downstream metabolic fates. Our findings highlight a previously unrecognized clearance mechanism for atypical sphingolipids with relevance to metabolic disease.

biochemistry↗

bioRxiv2026

Sex differences in the genetic basis of human recombination within 190,000 parent-child pairs

Structured AbstractO_ST_ABSIntroductionC_ST_ABSSexual reproduction uses a specialized cell division called meiosis, in which a single round of DNA replication is followed by two cell divisions to create hap-loid gametes. Genetic recombination in meiosis assures faithful segregation of chromosomes and establishes patterns of genetic linkage and inheritance. Meiotic recombination is thus a fundamental genomic process that shapes major features of the genomic landscape, influences mutation, and creates genetic diversity. RationaleThe genetic basis of between-individual variation in the occurrence and genomic location of meiotic recombination events remains poorly understood in humans. In particular, we lack an understanding of whether the same DNA regions shape vari-ation in recombination rates across studies. We address this by curating 112,144 maternal and 78,653 paternal meiosis events (parent-child pairs) across cohorts from four countries, creating the largest dataset of its kind. ResultsWe identify 58 independent genomic regions associated with the occurrence and genomic location of recombination events, 24 of which are unique across phe-notypes. Of the 24 unique regions, 6 are novel and specific to female meiotic recombination harboring genes linked to ovarian function, a 7th novel region is linked to male recombination. Many regions have significant sex-dependent ef-fects across studies. We estimate the between-sex genetic correlation for meiotic recombination rate to be 0.374 (0.072 SD), which captures the extent to which the genetic factors influencing a trait in one sex also impact the same trait in the other. 30-40% of the individual-level variation in recombination rate is at-tributable to DNA markers, of which only 37.7% (2.7% SD) and 47.1% (3.8% SD) is attributable to previously known global modulators of meiotic recombi-nation, in females and males respectively. We create a polygenic predictor that significantly predicts recombination rates of both women and men after adjusting for age and year of birth. We find significant correlations between female meiotic recombination, reproductive outcomes and reproductive aging, after adjusting for study ascertainment. ConclusionsIn the largest meta-analysis to date of human meiotic recombination, we identify novel loci with sex-specific effects that influence male and female recombination rate across multiple cohorts.

genomics↗

bioRxiv2026

Mapping corpus callosum architecture: developmental, genetic, and cognitive correlates in youth

The corpus callosum (CC), the largest interhemispheric white-matter tract, plays a central role in higher-order cognitive functions and is frequently altered in neurodevelopmental and psychiatric conditions. However, most diffusion MRI studies rely on tract-averaged measures, which obscure spatially specific microstructural variation along the tract that may hold biologically and functionally meaningful information. This study aimed to provide a fine-grained characterisation of spatial variation in callosal structure and to determine its developmental, genetic, and cognitive correlates. This study leveraged multimodal data from the Philadelphia Neurodevelopmental Cohort (PNC), comprising 1342 participants aged 8-21 years. Diffusion MRI and tractography-based segmentation were used to extract fine-grained spatial variation in diffusion metrics along seven callosal tracts. A candidate gene approach targeted four genes (DCC, CDH2, AKT3, and GLI3) previously linked to the neurobiological mechanisms underlying CC formation. Five cognitive factors were derived from the Penn Computerized Neurocognitive Battery using factor analysis. A two-stage functional data analysis approach to examine global and local associations between along-tract diffusion metrics and age, candidate genetic variants, and behavioural outcomes. Results revealed distinct midline-to-cortical variations of age-related diffusion metrics change across callosal subdivisions. Frontal and parietal heteromodal callosal pathways showed pronounced distal-segment maturation (F = 13 - 23, p [≤] 2.8x10e-16), whereas posterior sensorimotor and occipital callosal tracts exhibited more stable age associations along their lengths (F = 4.2 - 4.3, p = 10e-3). Genetic variations in callosal axon-guidance genes (ROBO1, IQCJ-SCHIP1, NRP1 and DCC) were associated with spatial variation in callosal diffusion metrics, particularly in anterior (rostrum) and posterior callosal subdivisions (isthmus and splenium) (F = 4.29 - 18.61, p-values = 2.2e-16 - 1.4e-04). These regions correspond to early-forming callosal compartments, suggesting that prenatal axon- guidance mechanisms leave enduring spatial patterns on callosal organisation. Finally, spatial variation in callosal microstructure was significantly associated with behavioural performance (F = 2.9 - 21.3, P = 2.8x10e-16 - 0.04), with the strongest and most spatially heterogeneous effects observed for complex cognition and executive functioning. Across all analyses, functional data models generally outperformed tract-averaged linear models, supporting the value of explicitly preserving spatial variation along callosal tracts. Our findings converge on the CC as a spatially differentiated structure in which early genetic modulators and developmental constraints shape region- and segment-specific microstructural architecture that is behaviourally relevant through childhood and adolescence.

neuroscience↗

bioRxiv2025

Lipid transport is necessary for neocortical lamination

We previously described the Alkuraya-Ku[c]inskas syndrome, a disorder associated with biallelic variants in BLTP1 (bridge-like lipid transfer protein), a.k.a. KIAA1109. The majority of probands die perinatally with corpus callosum agenesis, ventriculomegaly and arthrogryposis. Homozygous ablation of mouse Bltp1 resulted in similar preweaning lethality. Here, we describe ten novel patients expanding the characterization of this syndrome at the mild end of the phenotypic spectrum. To model this syndrome, we engineered Emx1-Cre-mediated conditional knockouts (cKO) in which Bltp1 expression is only removed in cortical and hippocampal neurons. This restricted ablation of Bltp1 recapitulated the preweaning lethality observed in the constitutional knockouts, suggesting that lack of BLTP1 expression in neurons is sufficient to cause death. Homozygous cKO presented a complete agenesis of the corpus callosum, a smaller anterior commissure, a malformed hippocampus and a reduced thickness of the cortical plate with a complete lack of defined structural layers and absence of radial glial and intermediate neural progenitors and mature neurons. As BLTP1 was shown to be a barrel-shaped tube containing lipids, we compared the amount of lipid species in the cKO and their control littermates cortexes. We observed significant depletions of ether-linked phosphatidylethanolamines and triglycerides and accumulations of sphingomyelins and hexosylceramides in cKOs. Our results are consistent with the recent description of BLTP1 as a tubular protein that transports phospholipids between the endoplasmic reticulum and the plasma membrane. They suggest that non-vesicular lipid transport is essential for neocortical and cerebellar lamination. Consistent with a BLTP1 role in cortex development we show that heterozygous carriers of a BLTP1 truncation variant presented a decrease in peripheral cortical grey matter suggesting an autosomal dominant inheritance pattern beside the already described autosomal recessive.

genetics↗

bioRxiv2025

Integration of genetic evidence to identify approved drug targets

Drugs targeting genes supported by human genetic evidence are more likely to succeed in clinical trials. While previous approaches have benchmarked individual methods such as genome-wide association studies (GWAS), rare variant burden testing, and quantitative trait locus (QTL)-informed Mendelian randomization, it remains unclear how best to integrate these signals for drug target discovery. Here, we compared gene-prioritization strategies across 30 complex traits, evaluating their ability to recover approved drug targets compiled into lenient and moderate gold-standard sets from six curated databases. Gene-level association scores from GWAS, expression QTL, protein QTL, and exome-based analyses were integrated using five unsupervised approaches. Predictive performance was assessed with area under the receiver operating characteristic curve (AUROC) and enrichment-based statistics. Across traits, GWAS alone ranked known drug targets on average [~]652 ranks (3.42%) above random expectation, and the minimum-rank-based integration strategy provided an improvement of further [~]558 positions (2.93%), achieving the best AUROC in 23 of 30 traits. When comparing genetic correlation and drug target overlap across trait pairs, we observed a significant positive association (r = 0.193; p = 5.46e-5), and cross-trait analyses further revealed that prioritization scores derived from related diseases could at times equal or even surpass a traits own performance. For instance, coronary artery disease data improved the prediction of stroke targets (p = 0.004), while inflammatory bowel disease data enhanced the prioritization of chronic kidney disease targets (p = 0.014). Taken together, these results demonstrate that integrating complementary genetic signals through a minimum-rank-based framework, combined with information from genetically related traits, systematically strengthens drug target identification across complex diseases.

genetics↗

bioRxiv2025

Parental haplotypes reconstruction in up to 440,209 individuals reveals recent assortative mating dynamics

Assortative mating (AM), the tendency for individuals to choose partners with similar traits, plays an important role in social stratification and has a wide-spread impact on the genetic architecture of complex human traits. A genetic footprint of this behaviour, genetic assortative mating (GAM), has been documented for many traits. However, most existing approaches to estimate GAM either rely on genotyped couples - rare in large biobanks - or on estimates based on gametic phase disequilibrium (GPD), which reflect cumulative effects over multiple generations and cannot detect short-term events. We introduce a novel, scalable approach that infers genetic data for the parental generation to boost statistical power and improve interpretability when estimating GAM in biobank cohorts. Our method reconstructs the two parental haplotypes of biobank individuals by leveraging state-of-the-art inter-chromosomal phasing based on close relative information. By correlating polygenic scores computed separately on maternally and paternally inherited haplotypes, we infer the extent of GAM. Applied to 245,884 individuals from the UK Biobank and 194,325 individuals from the Estonian Biobank, our haplotype-based estimates showed strong concordance with GAM estimates from genotyped mate pairs across 69 traits and improved performance compared to existing GPD-based methods. We replicated previously identified patterns of GAM for many traits (educational attainment, height, BMI, and alcohol consumption), and revealed new ones (overall health and sedentary lifestyle). Temporal and geographic stratification revealed accelerated assortment in recent generations -- particularly for education and height -- and modest differences between urban and rural contexts. Our method enables scalable, interpretable, and generation-specific estimation of GAM in large biobank cohorts, providing new insights into the dynamic nature of mate choice and its impact on the genetic makeup of populations.

genetics↗

bioRxiv2024

Associations between common genetic variants and income provide insights about the socioeconomic health gradient

We conducted a genome-wide association study (GWAS) on income among individuals of European descent and leveraged the results to investigate the socio-economic health gradient (N=668,288). We found 162 genomic loci associated with a common genetic factor underlying various income measures, all with small effect sizes. Our GWAS-derived polygenic index captures 1 - 4% of income variance, with only one-fourth attributed to direct genetic effects. A phenome-wide association study using this polygenic index showed reduced risks for a broad spectrum of diseases, including hypertension, obesity, type 2 diabetes, coronary atherosclerosis, depression, asthma, and back pain. The income factor showed a substantial genetic correlation (0.92, s.e. = .006) with educational attainment (EA). Accounting for EAs genetic overlap with income revealed that the remaining genetic signal for higher income related to better mental health but reduced physical health benefits and increased participation in risky behaviours such as drinking and smoking.

genetics↗

bioRxiv2023

Widespread natural selection on metabolite levels in humans

Natural selection acts ubiquitously on complex human traits, predominantly constraining the occurrence of extreme phenotypes (stabilizing selection). These constrains propagate to DNA sequence variants associated with traits under selection. The genetic imprints of such evolutionary events can thus be detected via combining effect size estimates from genetic association studies and the corresponding allele frequencies. While this approach has been successfully applied to high-level traits, the prevalence and mode of selection acting on molecular traits remains poorly understood. Here, we estimate the action of natural selection on genetic variants associated with metabolite levels, an important layer of molecular traits. By leveraging summary statistics of published genome-wide association studies with large sample sizes, we find strong evidence of stabilizing selection for 15 out of 97 plasma metabolites, with an overrepresentation of amino acids among such cases. Mendelian randomization analysis revealed that metabolites under stronger stabilizing selection display larger effects on key cardiometabolic traits, suggesting that maintaining a healthy cardiometabolic profile may be an important source of selective constraints on the metabolome. Metabolites under strong stabilizing selection in humans are also more conserved in their concentrations among diverse mammalian species, suggesting shared selective forces across micro and macroevolutionary time scales. Finally, we also found evidence for both disruptive and directional selection on specific lipid metabolites, potentially indicating ongoing evolutionary adaptation in humans. Overall, this study demonstrates that variation in metabolite levels among humans is frequently shaped by natural selection and this may be acting indirectly through maintaining cardiometabolic fitness.

genetics↗

bioRxiv2023

Incorporating discovery and replication GWAS into summary data Mendelian randomization studies: A review of current methods and a simple, general and powerful alternative

Mendelian Randomization (MR) is a popular method for using genetics to estimate the causal effect of a modifiable exposure on a health outcome. Single Nucleotide Polymorphisms (SNPs) are typically selected for inclusion if they pass a genome-wide significance threshold in order to guarantee that they are strong genetic instruments, but this also induces Winners curse, as SNP-exposure associations tend to be overestimated. In this paper, we consider how to combine SNP-exposure data from discovery and replication samples using two-sample and three-sample approaches to best account for Winners curse, weak instrument bias, and pleiotropy within a summary data MR framework, using only GWAS summary statistics. After reviewing several existing methods, that often correct for Winners curse at the individual SNP level, we propose a simple alternative based on the technique of regression calibration that enacts a global correction to the causal effect estimate directly. This approach does not only correct for Winners curse, but also simultaneously accounts for weak instruments bias. Regression calibration can be used with a wide range of existing MR methods, including pleiotropy-robust methods such as median-based and mode-based estimators. Extensive simulations and real data examples are used to illustrate the utility of the new approach. Software is provided for users to implement the method in practice. Author SummaryMendelian randomization is a method to explore causation in health research which exploits the random inheritance of genes from parents to offspring as a natural experiment. It attempts to quantify the effect of intervening and modifying a health exposure, such as a persons body mass, on a downstream outcome such as blood pressure. Causal estimates obtained using this method can be strongly influenced by the set of genes used, or more specifically, the rationale used to select them. For example, selecting only genes that are strongly associated with the health exposure can induce bias due to the Winners curse. Unfortunately, using genes with a small association can lead to so-called weak instrument bias leading to a no-win paradox. In this paper, we present a novel approach based on the technique of regression calibration to de-bias causal estimates in an MR study. Our approach relies on the use of two independent samples for the exposure (discovery and replication) to estimate the amount of bias that is expected for a specific set of genes, so that causal estimates can be re-calibrated accordingly. We use extensive simulations and applied examples to compare our approach to current methods and provide software for researchers to implement our approach in future studies.

genetics↗

bioRxiv2022

Causal Epigenetic Age Uncouples Damage and Adaptation

Machine learning models based on DNA methylation data can predict biological age but often lack causal insights. By harnessing large-scale genetic data through epigenome-wide Mendelian Randomization, we identified CpG sites potentially causal for aging-related traits. Neither the existing epigenetic clocks nor age-related differential DNA methylation are enriched in these sites. These CpGs include sites that contribute to aging and protect against it, yet their combined contribution negatively affects age-related traits. We established a novel framework to introduce causal information into epigenetic clocks, resulting in DamAge and AdaptAge--clocks that track detrimental and adaptive methylation changes, respectively. DamAge correlates with adverse outcomes, including mortality, while AdaptAge is associated with beneficial adaptations. These causality-enriched clocks exhibit sensitivity to short-term interventions. Our findings provide a detailed land-scape of CpG sites with putative causal links to lifespan and healthspan, facilitating the development of aging biomarkers, assessing interventions, and studying reversibility of age-associated changes.

genetics↗

bioRxiv2022

Correction for participation bias in the UK Biobank reveals non-negligible impact on genetic associations and downstream analyses

While large-scale volunteer-based studies such as the UK Biobank (UKBB) have become the cornerstone of genetic epidemiology, the study participants are rarely representative of their target population. Here, we aim to evaluate the impact of non-random participation in the UKBB, and to pin down areas of research that are particularly susceptible to biases when using non-representative samples for genome-wide discovery. By comparing 14 harmonized characteristics of the UKBB participants to that of a representative sample, we derived a model for participation probability. We then conducted inverse probability weighted genome-wide association analyses (wGWA) on 19 UKBB traits. Comparing the output obtained from wGWA (Neffective=94,643 - 102,215) to standard GWA analyses (N=263,464 - 283,749), we assessed the impact of participation bias on three estimated quantities, namely 1) genotype-phenotype associations, 2) heritability and genetic correlation estimates and 3) exposure-outcome causal effect estimates obtained from Mendelian Randomization. Participation bias can lead to both overestimation (e.g., cancer, education) and underestimation (e.g., coffee intake, depression/anxiety) of SNP effects. Novel SNPs were identified in wGWA for 12 of the included traits, highlighting SNPs missed as a result of participation bias. While the impact of participation bias on heritability estimates was small (average change in h2: 1.5%, maximum: 5%), substantial distortions were present for genetic correlations (average absolute change in rg: 0.07, maximum: 0.31) and Mendelian Randomization estimates (average absolute change in standardized estimates: 0.04, maximum: 0.15), most markedly for socio-behavioural traits including education, smoking and BMI. Overall, the bias mainly affected the magnitude of effects, rather than direction. In contrast, genome-wide findings for more molecular/physical traits (e.g., LDL, SBP) exhibited less bias as a result of selective participation. Our results highlight that participation bias can distort genomic findings obtained in non-representative samples, and we propose a viable solution to reduce such bias. Moving forward, more efforts ensuring either sample representativeness or correcting for participation bias are paramount, especially when investigating the genetic underpinnings of behaviour, lifestyles and social outcomes.

genetics↗

bioRxiv2022

Exploiting the mediating role of the metabolome to unravel transcript-to-phenotype associations

Despite the success of genome-wide association studies (GWASs) in identifying genetic variants associated with complex traits, understanding the mechanisms behind these statistical associations remains challenging. Several methods that integrate methylation, gene expression, and protein quantitative trait loci (QTLs) with GWAS data to determine their causal role in the path from genotype to phenotype have been proposed. Here, we developed and applied a multi-omics Mendelian randomization (MR) framework to study how metabolites mediate the effect of gene expression on complex traits. We identified 206 transcript-metabolite-trait causal triplets for 28 medically relevant phenotypes. Sixty-seven of these associations were missed by classical transcriptome-wide MR, which only uses gene expression and GWAS data. Among these, we identify biologically relevant pathways, such as between ANKH and calcium levels mediated by citrate and SLC6A12 and serum creatinine through modulation of the levels of the renal osmolyte betaine. We show that the signals missed by transcriptome-wide MR are found thanks to the gain in power allowed by integrating multiple omics-layer. Simulation analyses show that with larger molecular QTL studies and in case of mediated effects, our multi-omics MR framework outperforms classical MR approaches designed to detect causal relationships between single molecular traits and complex phenotypes.

genetics↗

bioRxiv2022

The contribution of mate-choice, couple convergence and confounding to assortative mating

Increased phenotypic similarity between partners, termed assortative mating (AM), has been observed for many traits. However, it is currently unclear if these observations are due to mate choice for certain phenotypes, post-mating convergence, or a result of confounding factors such as shared environment or indirect assortment. To dissect these underlying phenomena, we applied Mendelian randomisation (MR) to 51,664 couples in the UK biobank to a panel of 118 phenotypes under AM. We found that 54% (64 of 118) of the tested traits had a causal relationship between partners, with female-to-male effects on average being larger. Forty traits, including systolic blood pressure, basal metabolic rate, weight and height, showed significantly larger phenotypic correlation than MR-estimates, suggesting the presence of confounders. Subsequent analyses revealed household income, overall health rating, education and tobacco smoking as major overall confounders, accounting for 29.8, 14.1, 11.6, and 4.78%, of cross-partner phenotypic correlations, respectively. We detected limited evidence for couple-correlation convergence (e.g. increased similarity with respect to smoking and medication use), measured by stratifying couples by their time spent together. Finally, mediation analysis revealed that the vast majority (>77%) of causal associations between one trait of an individual and a different trait of their partner is indirect. For example, the causal effect of the BMI of an individual on the overall health rating of their partner is entirely acting through the BMI of their partner. In summary, this study revealed many novel causal effects within couples, shedding light on the impact of confounding on couple phenotypic similarity.

genetics↗

bioRxiv2022

Omics-informed CNV calls reduce false positive rate and improve power for CNV-trait associations

Copy number variations (CNV) are believed to play an important role in a wide range of complex traits but discovering such associations remains challenging. Whilst whole genome sequencing (WGS) is the gold standard approach for CNV detection, there are several orders of magnitude more samples with available genotyping microarray data. Such array data can be exploited for CNV detection using dedicated software (e.g., PennCNV), however these calls suffer from elevated false positive and negative rates. In this study, we developed a CNV quality score that weights PennCNV calls (pCNV) based on their likelihood of being true positive. First, we established a measure of pCNV reliability by leveraging evidence from multiple omics data (WGS, transcriptomics and methylomics) obtained from the same samples. Next, we built a predictor of omics-confirmed pCNVs, termed omics-informed quality score (OQS), using only PennCNV software output parameters. Promisingly, OQS assigned to pCNVs detected in close family members was up to 35% higher than the OQS of pCNVs not carried by other relatives (P < 3.0-10-90), outperforming other scores. Finally, in an association study of four anthropometric traits in 89,516 Estonian Biobank samples, the use of OQS led to a relative increase in the trait variance explained by CNVs of up to 34% compared to raw pCNVs or previous quality scores. Overall, we put forward a flexible framework to improve any CNV detection method leveraging multi-omics evidence, applied it to improve PennCNV calls and demonstrated its utility by improving the statistical power for downstream association analyses.

bioinformatics↗

bioRxiv2021

Polynomial Mendelian Randomization reveals widespread non-linear causal effects in the UK Biobank

Causal inference is a critical step in improving our understanding of biological processes and Mendelian randomisation (MR) has emerged as one of the foremost methods to efficiently interrogate diverse hypotheses using large-scale, observational data from biobanks. Although many extensions have been developed to address the three core assumptions of MR-based causal inference (relevance, exclusion restriction, and exchangeability), most approaches implicitly assume that any putative causal effect is linear. Here we propose PolyMR, an MR-based method which provides a polynomial approximation of an (arbitrary) causal function between an exposure and an outcome. We show that this method provides accurate inference of the shape and magnitude of causal functions with greater accuracy than existing methods. We applied this method to data from the UK Biobank, testing for effects between anthropometric traits and continuous health-related phenotypes and found most of these (84%) to have causal effects which deviate significantly from linear. These deviations ranged from slight attenuation at the extremes of the exposure distribution, to large changes in the magnitude of the effect across the range of the exposure (e.g. a 1 kg/m2 change in BMI having stronger effects on glucose levels if the initial BMI was higher), to non-monotonic causal relationships (e.g. the effects of BMI on cholesterol forming an inverted U shape). Finally, we show that the linearity assumption of the causal effect may lead to the misinterpretation of health risks at the individual level or heterogeneous effect estimates when using cohorts with differing average exposure levels.

genetics↗

bioRxiv2021

Parent-of-origin effects in the UK Biobank

Identical genetic variations can have different phenotypic effects depending on their parent of origin (PofO). Yet, studies focussing on PofO effects have been largely limited in terms of sample size due to the need of parental genomes or known genealogies. Here, we used a novel probabilistic approach to infer PofO of individual alleles in the UK Biobank that does not require parental genomes nor prior knowledge of genealogy. Our model uses Identity-By-Descent (IBD) sharing with second- and third-degree relatives to assign alleles to parental groups and leverages chromosome X data in males to distinguish maternal from paternal groups. When combined with robust haplotype inference and haploid imputation, this allowed us to infer the PofO at 5.4 million variants genome-wide for 26,393 UK Biobank individuals. We used this large dataset to systematically screen 59 biomarkers and 38 anthropomorphic phenotypes for PofO effects and discovered 101 significant associations, demonstrating that this type of effects contributes to the genetics of complex traits. Notably, we retrieved well known PofO effects, such as the MEG3/DLK1 locus on platelet count, and we discovered many new ones at loci often unsuspected of being imprinted and, in some cases, previously thought to harbour additive associations.

genomics↗