Individual Genotype Data Research Articles

Fast estimation of genetic correlation for biobank-scale data

SummaryGenetic correlation is an important parameter in efforts to understand the relationships among complex traits. Current methods that analyze individual genotype data for estimating genetic correlation are challenging to scale to large datasets. Methods that analyze summary data, while being computationally efficient, tend to yield estimates of genetic correlation with reduced precision. We propose SCORE (scalable genetic correlation estimator), a randomized method of moments estimator of genetic correlation that is both scalable and accurate. SCORE obtains more precise estimates of genetic correlations relative to summary-statistic methods that can be applied at scale; it achieves a reduction in standard error relative to LD-score regression (LDSC) and a reduction relative to high-definition likelihood (HDL) (averaged over all simulations). The efficiency of SCORE enables computation of genetic correlations on the UK Biobank dataset, consisting of K individuals and K SNPs, in a few h (orders of magnitude faster than methods that analyze individual data, such as GCTA). Across 780 pairs of traits in unrelated white British individuals in the UK Biobank, SCORE identifies significant genetic correlation between 200 additional pairs of traits over LDSC (beyond the 245 pairs identified by both).

Gut microbiota is associated with bone mineral density : an observational and genome-wide environmental interaction analysis in the UK Biobank cohort.

AimsDespite the interest in the association of gut microbiota with bone health, limited population-based studies of gut microbiota and bone mineral density (BMD) have been made. Our aim is to explore the possible association between gut microbiota and BMD.MethodsA total of 3,321 independent loci of gut microbiota were used to calculate the individual polygenic risk score (PRS) for 114 gut microbiota-related traits. The individual genotype data were obtained from UK Biobank cohort. Linear regressions were then conducted to evaluate the possible association of gut microbiota with L1-L4 BMD (n = 4,070), total BMD (n = 4,056), and femur total BMD (n = 4,054), respectively. PLINK 2.0 was used to detect the single-nucleotide polymorphism (SNP) × gut microbiota interaction effect on the risks of L1-L4 BMD, total BMD, and femur total BMD, respectively.ResultsWe detected five, three, and seven candidate gut microbiota-related traits for L1-L4 BMD, total BMD, and femur BMD, respectively, such as genus Dialister (p = 0.004) for L1-L4 BMD, and genus Eisenbergiella (p = 0.046) for total BMD. We also detected two common gut microbiota-related traits shared by L1-L4 BMD, total BMD, and femur total BMD, including genus Escherichia Shigella and genus Lactococcus. Interaction analysis of BMD detected several genes that interacted with gut microbiota, such as phospholipase D1 (PLD1) and endomucin (EMCN) interacting with genus Dialister in total BMD, and COL12A1 and Discs Large MAGUK Scaffold Protein 2 (DLG2) interacting with genus Lactococcus in femur BMD.ConclusionOur results suggest associations between gut microbiota and BMD, which will be helpful to further explore the regulation mechanism and intervention gut microbiota of BMD.Cite this article: Bone Joint Res 2021;10(11):734–741.

Vitamin D and the Risks of Depression and Anxiety: An Observational Analysis and Genome-Wide Environment Interaction Study.

Previous studies have suggested that vitamin D (VD) was associated with psychiatric diseases, but efforts to elucidate the functional relevance of VD with depression and anxiety from genetic perspective have been limited. Based on the UK Biobank cohort, we first calculated polygenic risk score (PRS) for VD from genome-wide association study (GWAS) data of VD. Linear and logistic regression analysis were conducted to evaluate the associations of VD traits with depression and anxiety traits, respectively. Then, using individual genotype and phenotype data from the UK Biobank, genome-wide environment interaction studies (GWEIS) were performed to identify the potential effects of gene × VD interactions on the risks of depression and anxiety traits. In the UK Biobank cohort, we observed significant associations of blood VD level with depression and anxiety traits, as well as significant associations of VD PRS and depression and anxiety traits. GWEIS identified multiple candidate loci, such as rs114086183 (p = 4.11 × 10−8, LRRTM4) for self-reported depression status and rs149760119 (p = 3.88 × 10−8, GNB5) for self-reported anxiety status. Our study results suggested that VD was negatively associated with depression and anxiety. GWEIS identified multiple candidate genes interacting with VD, providing novel clues for understanding the biological mechanism potential associations between VD and psychiatric disorders.

Assessing the effect of interaction between C-reactive protein and gut microbiome on the risks of anxiety and depression

Cumulative evidence shows that gut microbiome can influence brain function and behavior via the inflammatory processes. However, the role of interaction between gut dysbiosis and C-reactive protein (CRP) in the development of anxiety and depression remains to be elucidated. In this study, a total of 3321 independent single nucleotide polymorphism (SNP) loci associated with gut microbiome were driven from genome-wide association study (GWAS). Using individual level genotype data from UK Biobank, we then calculated the polygenetic risk scoring (PRS) of 114 gut microbiome related traits. Moreover, regression analysis was conducted to evaluate the possible effect of interaction between gut microbiome and CRP on the risks of Patient Health Questionnaire-9 (PHQ-9) (N = 113,693) and Generalized Anxiety Disorder-7 (GAD-7) (N = 114,219). At last, 11 candidate CRP × gut microbiome interaction with suggestive significance was detected for PHQ-9 score, such as F_Ruminococcaceae (β = − 0.009, P = 2.2 × 10–3), G_Akkermansia (β = − 0.008, P = 7.60 × 10–3), F_Acidaminococcaceae (β = 0.008, P = 1.22 × 10–2), G_Holdemanella (β = − 0.007, P = 1.39 × 10–2) and O_Lactobacillales (β = 0.006, P = 1.79× 10–2). 16 candidate CRP × gut microbiome interaction with suggestive significance was detected for GAD-7 score, such as O_Bacteroidales (β = 0.010, P = 4.00× 10–4), O_Selenomonadales (β = − 0.010, P = 1.20 × 10–3), O_Clostridiales (β = 0.009, P = 2.70 × 10–3) and G_Holdemanella (β = − 0.008, P = 4.20 × 10–3). Our results support the significant effect of interaction between CRP and gut microbiome on the risks of anxiety and depression, and identified several candidate gut microbiomes for them.

Genome-wide CNV investigation suggests a role for cadherin, Wnt, and p53 pathways in primary open-angle glaucoma

BackgroundTo investigate whether copy number variations (CNVs) are implicated in molecular mechanisms underlying primary open-angle glaucoma (POAG), we used genotype data of POAG individuals and healthy controls from two case-control studies, AGS (n = 278) and GLGS-UGLI (n = 1292). PennCNV, QuantiSNP, and cnvPartition programs were used to detect CNV. Stringent quality controls at both sample and marker levels were applied. The identified CNVs were intersected in CNV region (CNVR). After, we performed burden analysis, CNV-genome-wide association analysis, gene set overrepresentation and pathway analysis. In addition, in human eye tissues we assessed the expression of the genes lying within significant CNVRs.ResultsWe reported a statistically significant greater burden of CNVs in POAG cases compared to controls (p-value = 0,007). In common between the two cohorts, CNV-association analysis identified statistically significant CNVRs associated with POAG that span 11 genes (APC, BRCA2, COL3A1, HLA-DRB1, HLA-DRB5, HLA-DRB6, MFSD8, NIPBL, SCN1A, SDHB, and ZDHHC11). Functional annotation and pathway analysis suggested the involvement of cadherin, Wnt signalling, and p53 pathways.ConclusionsOur data suggest that CNVs may have a role in the susceptibility of POAG and they can reveal more information on the mechanism behind this disease. Additional genetic and functional studies are warranted to ascertain the contribution of CNVs in POAG.

Dietary Habit Is Associated with Depression and Intelligence: An Observational and Genome-Wide Environmental Interaction Analysis in the UK Biobank Cohort.

Dietary habits have considerable impact on brain development and mental health. Despite long-standing interest in the association of dietary habits with mental health, few population-based studies of dietary habits have assessed depression and fluid intelligence. Our aim is to investigate the association of dietary habits with depression and fluid intelligence. In total, 814 independent loci were utilized to calculate the individual polygenic risk score (PRS) for 143 dietary habit-related traits. The individual genotype data were obtained from the UK Biobank cohort. Regression analyses were then conducted to evaluate the association of dietary habits with depression and fluid intelligence, respectively. PLINK 2.0 was utilized to detect the single nucleotide polymorphism (SNP) × dietary habit interaction effect on the risks of depression and fluid intelligence. We detected 22 common dietary habit-related traits shared by depression and fluid intelligence, such as red wine glasses per month, and overall alcohol intake. For interaction analysis, we detected that OLFM1 interacted with champagne/white wine in depression, while SYNPO2 interacted with coffee type in fluid intelligence. Our study results provide novel useful information for understanding how eating habits affect the fluid intelligence and depression.

Fshr: a fish sex-determining locus shows variable incomplete penetrance across flathead grey mullet populations.

SummaryWhole-genome sequencing data were produced from a single flathead grey mullet female and assembled into a draft genome sequence, whereas publicly available sequence data were used to obtain a male draft sequence. Two pools, each consisting of 60 unrelated individuals, respectively, of male and female fish were analyzed using Pool-Sequencing. Mapping and analysis of Pool-Seq data against the draft genome(s) revealed >30 loci potentially associated with sex, the most promising locus of which, encoding the follicle-stimulating hormone receptor (fshr) and harboring two missense variants, was genotyped on 245 fish from four Mediterranean populations. Genotype data showed that fshr represents a previously unknown sex-determining locus, although the incomplete association pattern between fshr genotype and sex-phenotype, the variability of such pattern across different populations, and the presence of other candidate loci reveal that a greater complexity underlies sex determination in the flathead grey mullet.

Genotype-Based Gene Expression in Colon Tissue-Prediction Accuracy and Relationship with the Prognosis of Colorectal Cancer Patients.

Colorectal cancer (CRC) survival has environmental and inherited components. The expression of specific genes can be inferred based on individual genotypes—so called expression quantitative trait loci. In this study, we used the PrediXcan method to predict gene expression in normal colon tissue using individual genotype data from 91 CRC patients and examined the correlation ρ between predicted and measured gene expression levels. Out of 5434 predicted genes, 58% showed a negative ρ value and only 16% presented a ρ higher than 0.10. We subsequently investigated the association between genotype-based gene expression in colon tissue for genes with ρ > 0.10 and survival of 4436 CRC patients. We identified an inverse association between the predicted expression of ARID3B and CRC-specific survival for patients with a body mass index greater than or equal to 30 kg/m2 (HR (hazard ratio) = 0.66 for an expression higher vs. lower than the median, p = 0.005). This association was validated using genotype and clinical data from the UK Biobank (HR = 0.74, p = 0.04). In addition to the identification of ARID3B expression in normal colon tissue as a candidate prognostic biomarker for obese CRC patients, our study illustrates the challenges of genotype-based prediction of gene expression, and the advantage of reassessing the prediction accuracy in a subset of the study population using measured gene expression data.

Robust Huber-LASSO for improved prediction of protein, metabolite and gene expression levels relying on individual genotype data.

Least absolute shrinkage and selection operator (LASSO) regression is often applied to select the most promising set of single nucleotide polymorphisms (SNPs) associated with a molecular phenotype of interest. While the penalization parameter λ restricts the number of selected SNPs and the potential model overfitting, the least-squares loss function of standard LASSO regression translates into a strong dependence of statistical results on a small number of individuals with phenotypes or genotypes divergent from the majority of the study population—typically comprised of outliers and high-leverage observations.Robust methods have been developed to constrain the influence of divergent observations and generate statistical results that apply to the bulk of study data, but they have rarely been applied to genetic association studies. In this article, we review, for newcomers to the field of robust statistics, a novel version of standard LASSO that utilizes the Huber loss function. We conduct comprehensive simulations and analyze real protein, metabolite, mRNA expression and genotype data to compare the stability of penalization, the cross-iteration concordance of the model, the false-positive and true-positive rates and the prediction accuracy of standard and robust Huber-LASSO.Although the two methods showed controlled false-positive rates ≤2.1% and similar true-positive rates, robust Huber-LASSO outperformed standard LASSO in the accuracy of predicted protein, metabolite and gene expression levels using individual SNP data. The conducted simulations and real-data analyses show that robust Huber-LASSO represents a valuable alternative to standard LASSO in genetic studies of molecular phenotypes.

Genotype-driven pharmacokinetic simulations of warfarin levels in Puerto Ricans.

Objectives The inter-individual variability of warfarin dosing has been linked to genetic polymorphisms. This study was aimed at performing genotype-driven pharmacokinetic (PK) simulations to predict warfarin levels in Puerto Ricans. Methods Analysis of each individual dataset was performed by one-compartmental modeling using WinNonlin®v6.4. The k e of warfarin given a cytochrome P450 2C9 (CYP2C9) genotype ranged from 0.0189 to 0.0075h-1. K a and V d parameters were taken from literature. Data from 128 subjects were divided into two groups (i.e.,wild-types and carriers) and statistical analyses of PK parameters were performed by unpaired t-tests. Results In the carrier group (n=64), 53 subjects were single-carriers and 11 double-carriers (i.e.,*2/*2, *2/*3, *2/*5, *3/*5, and *3/*8). The mean peak concentration (Cmax) was higher for wild-type (0.36±0.12 vs. 0.32±0.14mg/L). Likewise, the average clearance (CL) parameter was faster among non-carriers (0.22±0.03 vs. 0.17±0.05L/h; p=0.0001), with also lower area under the curve (AUC) when compared to carriers (20.43±6.97 vs. 24.78±11.26h mg/L; p=0.025). Statistical analysis revealed a significant difference between groups with regard to AUC and CL, but not for Cmax. This can be explained by the variation of k e across different genotypes. Conclusions The results provided useful information for warfarin dosing predictions that take into consideration important individual PK and genotyping data.

Genotype-driven pharmacokinetic simulations of warfarin levels in Puerto Ricans.

The inter-individual variability of warfarin dosing has been linked to genetic polymorphisms. This study was aimed at performing genotype-driven pharmacokinetic (PK) simulations to predict warfarin levels in Puerto Ricans. Analysis of each individual dataset was performed by one-compartmental modeling using WinNonlin®v6.4. The ke of warfarin given a cytochrome P450 2C9 (CYP2C9) genotype ranged from 0.0189 to 0.0075h-1. Ka and Vd parameters were taken from literature. Data from 128 subjects were divided into two groups (i.e.,wild-types and carriers) and statistical analyses of PK parameters were performed by unpaired t-tests. In the carrier group (n=64), 53 subjects were single-carriers and 11 double-carriers (i.e.,*2/*2, *2/*3, *2/*5, *3/*5, and *3/*8). The mean peak concentration (Cmax) was higher for wild-type (0.36±0.12 vs. 0.32±0.14mg/L). Likewise, the average clearance (CL) parameter was faster among non-carriers (0.22±0.03 vs. 0.17±0.05L/h; p=0.0001), with also lower area under the curve (AUC) when compared to carriers (20.43±6.97 vs. 24.78±11.26h mg/L; p=0.025). Statistical analysis revealed a significant difference between groups with regard to AUC and CL, but not for Cmax. This can be explained by the variation of ke across different genotypes. The results provided useful information for warfarin dosing predictions that take into consideration important individual PK and genotyping data.

Association between herpes simplex virus 1 exposure and the risk of depression in UK Biobank.

BackgroundHerpes simplex virus‐1 (HSV‐1) infection is reported to be associated with depression. But limited efforts were made to investigate the relationship between HSV‐1 infection and the risk of depression, especially from the genetic perspective.MethodsIn UK Biobank cohort, linear and logistic regression analyses were first performed to test the association of HSV‐1 seropositivity/antibody with depression, including depression status (N = 2951) and Patient Health Questionnaire (PHQ) score (N = 2839). Using individual genotypic and phenotypic data from the UK Biobank, genome‐wide environmental interaction study (GWEIS) was then conducted by PLINK2.0 to evaluate gene × HSV‐1 interacting effect on the risk of depression. Finally, gene set enrichment analysis was conducted to identify the biological pathways involved in the observed gene × HSV‐1 interaction for depression.ResultIn UK Biobank cohort, significant associations were observed between depression status and HSV‐1 (odds ratio [OR] = 1.09; 95% confidence interval [CI], 1.02‐1.16; P = 2.40 × 10−2 for HSV‐1 antibody and OR = 1.28; 95% CI, 1.12‐1.47, P = 2.59 × 10−3 for HSV‐1 seropositivity). GWEIS revealed four significant gene × HSV‐1 interaction signals for PHQ score (all P < 5.0 × 10−8) and the leading loci was SULF2 (rs6094791, P = 8.60 × 10−9). Pathway analyses identified 21 pathways for PHQ score and 19 for depression status, including multiple neural development‐ and immune‐related ones, such as KEGG_NEUROACTIVE_LIGAND_RECEPTOR_INTERACTION (false discovery rate [FDR] = 3.18 × 10−2) for depression and LU_AGING_BRAIN_UP (FDR = 4.21 × 10−2) for PHQ score.ConclusionOur results suggested that HSV‐1 was associated with the risk of depression, which was modulated by the several genes that were related to the nerve development or immune function.

What is creating the height premium? New evidence from a Mendelian randomization analysis in China.

This study uses a Mendelian randomization approach to resolve the difficulties of identifying the causal relationship between height and earnings by using a unique sample of 3,427 respondents from mainland China with sociodemographic information linked to individual genotyping data. Exploiting genetic variations to create instrumental variables for observed height, we find that while OLS regressions yield that an additional centimeter in height is associated with a 10–13% increase in one’s annual earnings, IV estimates reveal only an insubstantial causal effect of height. Further analyses suggest that the observed height premium is likely to pick up the impacts of several cognitive/noncognitive skills on earnings confounded in previous studies, such as mental health, risk preference, and personality factors. Our study is the first empirical study that employs genetic IVs in developing countries, and our results contribute to the recent debate on the mechanism of height premium.

Tissue specific regulation of transcription in endometrium and association with disease.

STUDY QUESTIONAre genetic effects on endometrial gene expression tissue specific and/or associated with reproductive traits and diseases?SUMMARY ANSWERAnalyses of RNA-sequence data and individual genotype data from the endometrium identified novel and disease associated, genetic mechanisms regulating gene expression in the endometrium and showed evidence that these mechanisms are shared across biologically similar tissues.WHAT IS KNOWN ALREADYThe endometrium is a complex tissue vital for female reproduction and is a hypothesized source of cells initiating endometriosis. Understanding genetic regulation specific to, and shared between, tissue types can aid the identification of genes involved in complex genetic diseases.STUDY DESIGN, SIZE, DURATIONRNA-sequence and genotype data from 206 individuals was analysed and results were compared with large publicly available datasets.PARTICIPANTS/MATERIALS, SETTING, METHODSRNA-sequencing and genotype data from 206 endometrial samples was used to identify the influence of genetic variants on gene expression, via expression quantitative trait loci (eQTL) analysis and to compare these endometrial eQTLs with those in other tissues. To investigate the association between endometrial gene expression regulation and reproductive traits and diseases, we conducted a tissue enrichment analysis, transcriptome-wide association study (TWAS) and summary data-based Mendelian randomisation (SMR) analyses. Transcriptomic data was used to test differential gene expression between women with and without endometriosis.MAIN RESULTS AND THE ROLE OF CHANCEA tissue enrichment analysis with endometriosis genome-wide association study summary statistics showed that genes surrounding endometriosis risk loci were significantly enriched in reproductive tissues. A total of 444 sentinel cis-eQTLs (P < 2.57 × 10−9) and 30 trans-eQTLs (P < 4.65 × 10−13) were detected, including 327 novel cis-eQTLs in endometrium. A large proportion (85%) of endometrial eQTLs are present in other tissues. Genetic effects on endometrial gene expression were highly correlated with the genetic effects on reproductive (e.g. uterus, ovary) and digestive tissues (e.g. salivary gland, stomach), supporting a shared genetic regulation of gene expression in biologically similar tissues. The TWAS analysis indicated that gene expression at 39 loci is associated with endometriosis, including five known endometriosis risk loci. SMR analyses identified potential target genes pleiotropically or causally associated with reproductive traits and diseases including endometriosis. However, without taking account of genetic variants, a direct comparison between women with and without endometriosis showed no significant difference in endometrial gene expression.LARGE SCALE DATAThe eQTL dataset generated in this study is available at http://reproductivegenomics.com.au/shiny/endo_eqtl_rna/. Additional datasets supporting the conclusions of this article are included within the article and the supplementary information files, or are available on reasonable request.LIMITATIONS, REASONS FOR CAUTIONData are derived from fresh tissue samples and expression levels are an average of expression from different cell types within the endometrium. Subtle cell-specifc expression changes may not be detected and differences in cell composition between samples and across the menstrual cycle will contribute to sample variability. Power to detect tissue specific eQTLs and differences between women with and without endometriosis was limited by the sample size in this study. The statistical approaches used in this study identify the likely gene targets for specific genetic risk factors, but not the functional mechanism by which changes in gene expression may influence disease risk.WIDER IMPLICATIONS OF THE FINDINGSOur results identify novel genetic variants that regulate gene expression in endometrium and the majority of these are shared across tissues. This allows analysis with large publicly available datasets to identify targets for female reproductive traits and diseases. Much larger studies will be required to identify genetic regulation of gene expression that will be specific to endometrium.STUDY FUNDING/COMPETING INTEREST(S)This work was supported by the National Health and Medical Research Council (NHMRC) under project grants GNT1026033, GNT1049472, GNT1046880, GNT1050208, GNT1105321, GNT1083405 and GNT1107258. G.W.M is supported by a NHMRC Fellowship (GNT1078399). J.Y is supported by an ARC Fellowship (FT180100186). There are no competing interests.

Pharmacogenomics landscape of COVID-19 therapy response in Serbian population and comparison with worldwide populations.

Since there are no certified therapeutics to treat COVID-19 patients, drug repurposing became important. With lack of time to test individual pharmacogenomics markers, population pharmacogenomics could be helpful in predicting a higher risk of developing adverse reactions and treatment failure in COVID-19 patients. Aim of our study was to identify pharmacogenes and pharmacogenomics markers associated with drugs recommended for COVID-19 treatment, chloroquine/hydroxychloroquine, azithromycin, lopinavir and ritonavir, in population of Serbia and other world populations. Genotype information of 143 individuals of Serbian origin was extracted from database previously obtained using TruSight One Gene Panel (Illumina). Genotype data of individuals from different world populations were extracted from the 1000 Genome Project. Fisher's exact test was used for comparison of allele frequencies. We have identified 11 potential pharmacogenomics markers in 7 pharmacogenes relevant for COVID-19 treatment. Based on high alternative allele frequencies in population and the functional effect of the variants, ABCB1 rs1045642 and rs2032582 could be relevant for reduced clearance of azithromycin, lopinavir and ritonavir drugs and UGT1A7 rs17868323 for hyperbilirubinemia in ritonavir treated COVID-19 patients in Serbian population. SLCO1B1 rs4149056 is a potential marker of lopinavir response, especially in Italian population. Our results confirmed that pharmacogenomics profile of African population is different from the rest of the world. Considering population specific pharmacogenomics landscape, preemptive testing for pharmacogenes relevant for drugs used in COVID-19 treatment could contribute to better understanding of the inconsistency in therapy response and could be applied to improve the outcome of the COVID-19 patients.

Creation of an avocado unambiguous genotype SNP database for germplasm curation and as an aid to breeders

Avocado (Persea americana) is an important tropical and subtropical fruit tree crop. Traditional tree breeding programs face the challenges of long generation times and significant expense in land and personnel resources. Avocado selection and breeding can be more efficient and less expensive through the development and application of molecular markers. A total of 1524 individuals were genotyped with 384 SNPs creating the largest SNP genotype database for avocado. These individuals correspond to four extensive germplasm collections including two housed in Florida and two in California. In addition, hybrids and selections from two rootstock breeding programs have been genotyped. Genotype data were analyzed using an affinity propagation method to define 155 groups. The 384 SNP markers provided accurate genotype data for individuals from different Persea species as well as half-siblings. Therefore, the majority of the genetic diversity of the avocado germplasm and related species that were genotyped has been captured. A simple visual method can also be used to identify self-pollinated individuals among the half-siblings of known maternal parents and, in some cases, to infer likely candidates for the paternal parent. Finally, this dataset is unambiguous so breeders can determine the genetic diversity of their breeding stock to optimize avocado breeding and selection programs by identifying outcrossed individuals at the seedling stage, thus increasing the efficiency of avocado genetic improvement.

Genetic regulation of methylation in human endometrium and blood and gene targets for reproductive diseases

BackgroundMajor challenges in understanding the functional consequences of genetic risk factors for human disease are which tissues and cell types are affected and the limited availability of suitable tissue. The aim of this study was to evaluate tissue-specific genotype-epigenetic characteristics in DNA samples from both endometrium and blood collected from women at different stages of the menstrual cycle and relate results to genetic risk factors for reproductive traits and diseases.ResultsWe analysed DNA methylation (DNAm) data from endometrium and blood samples from 66 European women. Methylation profiles were compared between stages of the menstrual cycle, and changes in methylation overlaid with changes in transcription and genotypes. We observed large changes in methylation (27,262 DNAm probes) across the menstrual cycle in endometrium that were not observed in blood. Individual genotype data was tested for association with methylation at 443,016 and 443,101 DNAm probes in endometrium and blood respectively to identify methylation quantitative trait loci (mQTLs). A total of 4546 sentinel cis-mQTLs (P < 1.13 × 10−10) and 434 sentinel trans-mQTLs (P < 2.29 × 10−12) were detected in endometrium and 6615 sentinel cis-mQTLs (P < 1.13 × 10−10) and 590 sentinel trans-mQTLs (P < 2.29 × 10−12) were detected in blood. Following secondary analyses, conducted to test for overlap between mQTLs in the two tissues, we found that 62% of endometrial cis-mQTLs were also observed in blood and the genetic effects between tissues were highly correlated. A number of mQTL SNPs were associated with reproductive traits and diseases, including one mQTL located in a known risk region for endometriosis (near GREB1).ConclusionsWe report novel findings characterising genetic regulation of methylation in endometrium and the association of endometrial mQTLs with endometriosis risk and other reproductive traits and diseases. The high correlation of genetic effects between tissues highlights the potential to exploit the power of large mQTL datasets in endometrial research and identify target genes for functional studies. However, tissue-specific methylation profiles and genetic effects also highlight the importance of also using disease-relevant tissues when investigating molecular mechanisms of disease risk.

Gene-based association tests using GWAS summary statistics.

A huge number of genome-wide association studies (GWAS) summary statistics freely available in databases provide a new material for gene-based association analysis aimed at identifying rare genetic variants. Only a few of the many popular gene-based methods developed for individual genotype and phenotype data are adapted for the practical use of the GWAS summary statistics as input. We analytically prove and numerically illustrate that all popular powerful methods developed for gene-based association analysis of individual phenotype and genotype data can be modified to utilize GWAS summary statistics. We have modified and implemented all of the popular methods, including burden and kernel machine-based tests, multiple and functional linear regression, principal components analysis and others, in the R package sumFREGAT. Using real summary statistics for coronary artery disease, we show that the new package is able to detect genes not found by the existing packages. The R package sumFREGAT is freely and publicly available at: https://CRAN.R-project.org/package=sumFREGAT. Supplementary data are available at Bioinformatics online.

Using transcriptome sequencing and pooled exome capture to study local adaptation in the giga-genome of Pinus cembra.

Despite decreasing sequencing costs, whole-genome sequencing for population-based genome scans for selection is still prohibitively expensive for organisms with large genomes. Moreover, the repetitive nature of large genomes often represents a challenge in bioinformatic and downstream analyses. Here, we use in-depth transcriptome sequencing to design probes for exome capture in Swiss stone pine (Pinus cembra), a conifer with an estimated genome size of 29.3 Gbp and no reference genome available. We successfully applied around 55,000 self-designed probes, targeting 25,000 contigs, to DNA pools of seven populations from the Swiss Alps and identified >160,000 SNPs in around 15,000 contigs. The probes performed equally well in pools of the closely related species Pinus sibirica; in both species, more than 70% of the targeted contigs were sequenced at a depth ≥40× (number of haplotypes in the pool). However, a thorough analysis of individually sequenced P.cembra samples indicated that a majority of the contigs (63%) represented multi-copy genes. We therefore removed paralogous contigs based on heterozygote excess and deviation from allele balance. Without putatively paralogous contigs, allele frequencies of population pools represented accurate estimates of individually determined allele frequencies. We show that inferences of neutral and adaptive genetic variation may be biased when not accounting for such multi-copy genes. Without individual genotype data, it would have been nearly impossible to recognize and deal with the problem of multi-copy contigs. We advocate to put more emphasis on identifying paralogous loci, which will be facilitated by the establishment of additional high-quality reference genomes.

SU18 - INVESTIGATING SHRINKAGE METHODS TO IMPROVE ACCURACY OF GWAS AND PRS EFFECT SIZE ESTIMATES

Background As the scale and power of GWAS have increased to detect the small genetic effect sizes involved in complex polygenic traits, the results of GWAS, especially SNP effect size estimates, are increasingly utilised for prediction. A popular approach is the application of Polygenic Risk Scores (PRS). However, the effect size estimates are inherently prone to overfit the specific samples on which the GWAS were performed (e.g. “the Winner's Curse”). The inflation of effect size estimation reduces the out-of-sample accuracy of prediction based on GWAS results. Many shrinkage methods have been developed to correct for such inflation. As GWAS results and large individual genotype data sets become widely available, there is greater opportunity to accurately evaluate effect size inflation and compare the performance of different shrinkage methods. Methods We develop a novel permutation-based shrinkage method and compare its performance with previously developed methods based on local false discovery rate and the LASSO. We evaluate the performance of the methods across a wide range of phenotypes and investigate different factors (e.g. phenotype heritability) that influence their impact on the predictive power of polygenic risk scores. Results Using large-scale real genotype data, we show that SNP effect sizes are markedly overfit even in relatively large samples, but that our permutation-based shrinkage approach can improve PRS prediction dramatically. Discussion Our results suggest that GWAS results can be adjusted using an efficient empirical approach to provide more accurate effect size estimates and thus greater downstream predictive power. This approach could be applied to a wide variety of big data settings.