GXwasR: A Toolkit for Investigating Sex-Differentiated Genetic Effects on Complex Traits

GXwasR: A Toolkit for Investigating Sex-Differentiated Genetic Effects on Complex Traits | medRxiv /* */ /* */ <!-- <!-- /*! * yepnope1.5.4 * (c) WTFPL, GPLv2 */ (function(a,b,c){function d(a){return"[object Function]"==o.call(a)}function e(a){return"string"==typeof a}function f(){}function g(a){return!a||"loaded"==a||"complete"==a||"uninitialized"==a}function h(){var a=p.shift();q=1,a?a.t?m(function(){("c"==a.t?B.injectCss:B.injectJs)(a.s,0,a.a,a.x,a.e,1)},0):(a(),h()):q=0}function i(a,c,d,e,f,i,j){function k(b){if(!o&&g(l.readyState)&&(u.r=o=1,!q&&h(),l.onload=l.onreadystatechange=null,b)){"img"!=a&&m(function(){t.removeChild(l)},50);for(var d in y[c])y[c].hasOwnProperty(d)&&y[c][d].onload()}}var j=j||B.errorTimeout,l=b.createElement(a),o=0,r=0,u={t:d,s:c,e:f,a:i,x:j};1===y[c]&&(r=1,y[c]=[]),"object"==a?l.data=c:(l.src=c,l.type=a),l.width=l.height="0",l.onerror=l.onload=l.onreadystatechange=function(){k.call(this,r)},p.splice(e,0,u),"img"!=a&&(r||2===y[c]?(t.insertBefore(l,s?null:n),m(k,j)):y[c].push(l))}function j(a,b,c,d,f){return q=0,b=b||"j",e(a)?i("c"==b?v:u,a,b,this.i++,c,d,f):(p.splice(this.i++,0,a),1==p.length&&h()),this}function k(){var a=B;return a.loader={load:j,i:0},a}var l=b.documentElement,m=a.setTimeout,n=b.getElementsByTagName("script")[0],o={}.toString,p=[],q=0,r="MozAppearance"in l.style,s=r&&!!b.createRange().compareNode,t=s?l:n.parentNode,l=a.opera&&"[object Opera]"==o.call(a.opera),l=!!b.attachEvent&&!l,u=r?"object":l?"script":"img",v=l?"script":u,w=Array.isArray||function(a){return"[object Array]"==o.call(a)},x=[],y={},z={timeout:function(a,b){return b.length&&(a.timeout=b[0]),a}},A,B;B=function(a){function b(a){var a=a.split("!"),b=x.length,c=a.pop(),d=a.length,c={url:c,origUrl:c,prefixes:a},e,f,g;for(f=0;f<d;f++)g=a[f].split("="),(e=z[g.shift()])&&(c=e(c,g));for(f=0;f<b;f++)c=x[f](c);return c}function g(a,e,f,g,h){var i=b(a),j=i.autoCallback;i.url.split(".").pop().split("?").shift(),i.bypass||(e&&(e=d(e)?e:e[a]||e[g]||e[a.split("/").pop().split("?")[0]]),i.instead?i.instead(a,e,f,g,h):(y[i.url]?i.noexec=!0:y[i.url]=1,f.load(i.url,i.forceCSS||!i.forceJS&&"css"==i.url.split(".").pop().split("?").shift()?"c":c,i.noexec,i.attrs,i.timeout),(d(e)||d(j))&&f.load(function(){k(),e&&e(i.origUrl,h,g),j&&j(i.origUrl,h,g),y[i.url]=2})))}function h(a,b){function c(a,c){if(a){if(e(a))c||(j=function(){var a=[].slice.call(arguments);k.apply(this,a),l()}),g(a,j,b,0,h);else if(Object(a)===a)for(n in m=function(){var b=0,c;for(c in a)a.hasOwnProperty(c)&&b++;return b}(),a)a.hasOwnProperty(n)&&(!c&&!--m&&(d(j)?j=function(){var a=[].slice.call(arguments);k.apply(this,a),l()}:j[n]=function(a){return function(){var b=[].slice.call(arguments);a&&a.apply(this,b),l()}}(k[n])),g(a[n],j,b,n,h))}else!c&&l()}var h=!!a.test,i=a.load||a.both,j=a.callback||f,k=j,l=a.complete||f,m,n;c(h?a.yep:a.nope,!!i),i&&c(i)}var i,j,l=this.yepnope.loader;if(e(a))g(a,0,l,0);else if(w(a))for(i=0;i (function(w,d,s,l,i){w[l]=w[l]||[];w[l].push({'gtm.start':new Date().getTime(),event:'gtm.js'});var f=d.getElementsByTagName(s)[0];var j=d.createElement(s);var dl=l!='dataLayer'?'&l='+l:'';j.src='//www.googletagmanager.com/gtm.js?id='+i+dl;j.type='text/javascript';j.async=true;f.parentNode.insertBefore(j,f);})(window,document,'script','dataLayer','GTM-P4HH5NV'); Skip to main content Home About Submit ALERTS / RSS Search for this keyword Advanced Search GXwasR: A Toolkit for Investigating Sex-Differentiated Genetic Effects on Complex Traits View ORCID Profile Banabithi Bose , Freida Blostein , View ORCID Profile Jeewoo Kim , View ORCID Profile Jessica Winters , View ORCID Profile Ky’Era V. Actkins , View ORCID Profile David Mayer , View ORCID Profile Harrsha Congivaram , View ORCID Profile Maria Niarchou , View ORCID Profile Digna Velez Edwards , View ORCID Profile Lea K. Davis , View ORCID Profile Barbara E. Stranger doi: https://doi.org/10.1101/2025.06.10.25329327 Banabithi Bose 1 Department of Biomedical Informatics, University of Colorado School of Medicine , Aurora, CO Find this author on Google Scholar Find this author on PubMed Search for this author on this site ORCID record for Banabithi Bose For correspondence: banabithi.bose{at}cuanschutz.edu barbara.stranger{at}cuanschutz.edu Freida Blostein 2 Vanderbilt Genetics Institute, Vanderbilt University Medical Center , Nashville, TN 3 Division of Genetic Medicine, Department of Medicine, Vanderbilt University Medical Center , Nashville, TN Find this author on Google Scholar Find this author on PubMed Search for this author on this site Jeewoo Kim 2 Vanderbilt Genetics Institute, Vanderbilt University Medical Center , Nashville, TN 3 Division of Genetic Medicine, Department of Medicine, Vanderbilt University Medical Center , Nashville, TN 4 Division of Quantitative and Clinical Sciences, Vanderbilt University , Nashville, TN 5 Medical Scientist Training Program, Vanderbilt University , Nashville, TN Find this author on Google Scholar Find this author on PubMed Search for this author on this site ORCID record for Jeewoo Kim Jessica Winters 2 Vanderbilt Genetics Institute, Vanderbilt University Medical Center , Nashville, TN 3 Division of Genetic Medicine, Department of Medicine, Vanderbilt University Medical Center , Nashville, TN Find this author on Google Scholar Find this author on PubMed Search for this author on this site ORCID record for Jessica Winters Ky’Era V. Actkins 2 Vanderbilt Genetics Institute, Vanderbilt University Medical Center , Nashville, TN 3 Division of Genetic Medicine, Department of Medicine, Vanderbilt University Medical Center , Nashville, TN Find this author on Google Scholar Find this author on PubMed Search for this author on this site ORCID record for Ky’Era V. Actkins David Mayer 1 Department of Biomedical Informatics, University of Colorado School of Medicine , Aurora, CO Find this author on Google Scholar Find this author on PubMed Search for this author on this site ORCID record for David Mayer Harrsha Congivaram 6 Center for Genetic Medicine, Feinberg School of Medicine, Northwestern University , Chicago, IL Find this author on Google Scholar Find this author on PubMed Search for this author on this site ORCID record for Harrsha Congivaram Maria Niarchou 3 Division of Genetic Medicine, Department of Medicine, Vanderbilt University Medical Center , Nashville, TN Find this author on Google Scholar Find this author on PubMed Search for this author on this site ORCID record for Maria Niarchou Digna Velez Edwards 4 Division of Quantitative and Clinical Sciences, Vanderbilt University , Nashville, TN 7 Department of Obstetrics and Gynecology, Department of Medicine, Vanderbilt University Medical Center , Nashville, TN 8 Department of Biomedical Informatics, Institute for Medicine and Public Health, Vanderbilt Epidemiology Center , Nashville, TN Find this author on Google Scholar Find this author on PubMed Search for this author on this site ORCID record for Digna Velez Edwards Lea K. Davis 2 Vanderbilt Genetics Institute, Vanderbilt University Medical Center , Nashville, TN 3 Division of Genetic Medicine, Department of Medicine, Vanderbilt University Medical Center , Nashville, TN 9 Mount Sinai Million Health Discoveries Program, Charles Bronfman Institute for Personalized Medicine , New York, NY 10 Department of AI and Human Health, Icahn School of Medicine at Mount Sinai , New York, NY 11 Division of Data-Driven and Digital Medicine, Department of Medicine, Icahn School of Medicine at Mount Sinai , New York, NY 12 New York Genome Center , New York, NY Find this author on Google Scholar Find this author on PubMed Search for this author on this site ORCID record for Lea K. Davis Barbara E. Stranger 1 Department of Biomedical Informatics, University of Colorado School of Medicine , Aurora, CO 6 Center for Genetic Medicine, Feinberg School of Medicine, Northwestern University , Chicago, IL Find this author on Google Scholar Find this author on PubMed Search for this author on this site ORCID record for Barbara E. Stranger For correspondence: banabithi.bose{at}cuanschutz.edu barbara.stranger{at}cuanschutz.edu Abstract Full Text Info/History Metrics Supplementary material Data/Code Preview PDF Abstract Many complex traits in humans exhibit sex differences. Recent advances in understanding context-specific genetic effects underscore the imperative for sex-aware genetic analyses to assess the role of sex in the genetic basis of human phenotypes. Despite increasing recognition of the importance of sex as a biological variable, current computational tools for genome-wide association studies (GWAS) lack functionality for performing nuanced sex-aware quality control and analysis. In response, we present GXwasR, a comprehensive R package that implements statistical genetic models for performing both GWAS and X-Chromosome-Wide Association Studies (XWAS). Designed for both sex-combined and sex-stratified analyses, GXwasR can account for unique features of chromosome X, including X-chromosome inactivation. Beyond enabling association analysis and meta-analysis, the tool has functions to a) test for sex differences in SNP-level genetic effects and heritability, b) assess genetic correlation, c) compute sex-aware polygenic risk scores (PRS), and d) support enhanced quality control protocols expressly designed for these purposes. GXwasR thus serves as a computational resource to enable investigations into the role of sex in the genetic basis of complex human phenotypes. Introduction In biomedical research, sex emerged as an important biological variable 1 , 2 , profoundly shaping human health and disease 3 , 4 . Advances in genetics and biomedicine are revealing how genetic and physiological sex differences interact to influence complex traits such as autoimmune and cardiometabolic diseases, cancer, and response to treatment. This underscores the importance of incorporating sex as a factor in biomedical research, aligning with precision medicine’s focus on individualized care. As a result, global research organizations now mandate the consideration of sex as a biological variable throughout all stages of biomedical research 4 . Yet despite these advances, our understanding of how sex affects the genetic architecture of complex traits remains incomplete. Although sex differences can substantially influence disease risk and trait variation 5 – 7 , most genome-wide association studies (GWAS) do not explicitly test for sex-specific effects—often omitting the sex chromosomes entirely 8 , 9 . This omission is especially concerning given the unique features of the X and Y chromosomes 10 – 14 , which likely harbor underexplored genetic variation. Addressing this knowledge gap calls for robust, sex-aware analytical tools capable of integrating both autosomal data and the distinct biology of the sex chromosomes. Such a tool must correctly account for the intricacies of both autosomal and Chromosome X (hereafter, ChrX) genetic data, accommodating unique aspects such as XCI patterns 9 , 15 , 16 , and be able to identify sex-differentiated genetic effects. We present GXwasR, an R-based tool enabling sex-aware genetic analysis across autosomes and chromosome X. Building on our recent guidelines for analyzing sex differences in complex traits 17 , GXwasR provides thirty-three functions organized into six main categories: (A) pre-imputation quality control (QC), (B) post-imputation QC, (C) sex-combined and sex-stratified GWAS and XWAS, (D) tests for sex-differentiated genetic effects, (E) additional analyses (heritability estimation, genetic correlation, gene-based testing, polygenic risk scores (PRS) computation, and meta-analysis), and (F) utility functions. Beyond QC and association testing, GXwasR supports additional analyses, including SNP × sex interaction tests and sex-stratified association followed by Z-score or t-statistic comparisons. It also addresses unique aspects of ChrX, such as random, skewed, and escaped X-chromosome inactivation (XCI) and pseudoautosomal regions (PAR)—and introduces sex-specific checks for minor allele frequency (MAF) and Hardy–Weinberg equilibrium (HWE). GXwasR integrates seamlessly with established tools, PLINK 18 ,GCTA 19 , REGENIE 20 and SAIGE 21 . It leverages R’s statistical and graphical capabilities to streamline sex-aware genetic research, offering visualization options, flexible workflows, and extensive documentation. Together these features make GXwasR a versatile resource for investigating sex differences in the genetics of complex traits. Sex-aware QC of genotype data with GXwasR GXwasR integrates a comprehensive QC pipeline for analyzing both autosomal and ChrX genotypes, covering pre- and post-imputation steps ( Figure 1 ). It can accommodate pre- and post-imputation genotype data as well as variants derived from whole-genome sequencing. Importantly, this tool is not limited to sex-aware analysis; it also accommodates combined-sex QC and analysis. Download figure Open in new tab Figure 1. Pipelines for sex-aware QC of A) pre-imputed and B) post-imputed genotypes using GXwasR functions. Download figure Open in new tab Figure 3: Overview of sex-aware GWAS and XWAS methods implemented in GXwasR: A) sex-combined tests using different ways of handling ChrX; B) sex-stratified tests; C) sex-differential Z-score test comparing effect sizes using standard errors; D) sex-differential t-test accounting for correlation between estimates; and E) methods to combine p-values from sex-stratified analyses using Fisher’s or Stouffer’s approaches. Download figure Open in new tab Figure 4: Overview of the PRS workflow, including: QC steps for discovery and target datasets (e.g., SNP info score filtering, removal of indels and ambiguous SNPs); sex-combined or sex-specific PRS construction involving LD clumping, p-value thresholding, PRS computation, and model fitting; and selection of the best-fit threshold using R-square metrics, with sex-stratified analyses indicated by asterisks. The pre-imputation QC pipeline is supported by the following features: Ancestry and chromosomal sex verification The tool verifies user-provided ancestry labels or determines them using genetic data from reference populations. It also confirms or imputes sex assignments using genetic data from sex chromosomes. SNP level QC The tool performs QC of autosomal and ChrX variants. This involves filtering based on quality metrics, including genotyping efficiency, call rate, HWE, MAF and high-LD regions. Sample level QC The tool can identify and remove samples with a high rate of missing data and or heterozygosity. Flexible thresholds While default thresholds are provided based on prior literature 22 , 23 , but users can adjust them as needed. Visual displays The tool provides visualization, including plots and tables. Detailed GXwasR pipelines and tutorials for pre-imputation QC, ancestry estimation and post-imputation QC are available in the Supplementary File . The post-imputation QC pipeline is supported by the following features ( Figure 1 ): Filtering SNPs with low imputation accuracy The tool can remove SNPs based on r 2 values and average certainty of best-guess genotypes. The tool can either remove or flag specific regions of ChrX that may confound the analysis, or integrate these regions with autosomes if appropriate The tool can exclude or flag specific regions of ChrX that could potentially confound the analysis, (i.e., Pseudoautosomal Regions 24 , X-Transposed Regions 25 , and Ampliconic regions 26 ). Sex differences in allele frequencies within controls GXwasR can test for sex-differentiated allele frequencies among controls which can indicate technical issues or biases within the cohort. MAF thresholds for autosomal SNPs The tool can filter autosomal SNPs based on MAF, which can be adjusted separately for males and females. Quality control for ChrX The tool implements sex-stratified QC for ChrX variants while addressing the unique characteristics of ChrX. Post-imputation Sample QC The tool can remove samples based on missing genotype rate, heterozygosity, and identity-by-descent (IBD; pi-hat) 17 separately for each sex. Sex-aware genetic association analysis GXwasR performs sex-aware genetic association studies while accounting for unique aspects of sex chromosomes and autosomes. It includes the GXwas() function for full genome-wide analysis (GWAS + XWAS), with special modeling of ChrX variants. Unlike autosomes where genotypes are coded as 0, 1, and 2, ChrX variants can be subject to one of several XCI patterns. For XWAS, GXwasR codes female ChrX genotypes as 0, 1, and 2, while male genotypes are coded to reflect dosage compensation nuances, supporting several different XCI patterns (Figure 2). The models supported by GXwas() for running GWAS including XWAS are FMcombx01 Sex-combined analysis of autosomes and sex chromosomes, where female ChrX genotypes are coded as autosomal variants (i.e., 0, 1, and 2), whereas male ChrX genotypes are coded as 0 and 1 to reflect the XCI-Random pattern 27 . FMcombx02 Sex-combined analysis of autosomes and sex chromosomes, where female ChrX genotypes are coded as autosomal variants, whereas male ChrX genotypes are coded as 0 and 2, to reflect the XCI-Escape pattern 28 . Both models can accommodate covariates and can perform interaction tests, such as genotype-by-sex (i.e, SNP × Sex). Fmstratified Sex-stratified GWAS and XWAS with covariates. Fmcomb Combining the p-values of the FMstratified XWAS tests to obtain a final sex-combined XWAS p-values for each ChrX SNP using Fisher method 29 , Fisher method with permutation 30 and Stouffer method 31 to allow the ChrX SNP tested to have different, even an opposite, effect on disease risk in males and females. This functionality is automated into the FMstratified model of GXwas() function as well. GWAcxci This model represents a female-male combined analysis for autosomes, akin to FMcombx01 or FMcombx02 . However, for ChrX, it incorporates a continuous XCI pattern 32 , particularly suitable for binary traits. The GXwas() function automatically generates Manhattan and quantile-quantile (QQ) plots to visualize GWAS and XWAS results. Test for sex-differentiated genetic architecture GXwasR can test for sex-differentiated genetic effects using sex-stratified GWAS summary statistics as input. For each SNP, the tool utilizes a t-test and Z-score to test for a sex difference in effect size (Figure 2). For the t-test, the SexDiff() function is applied to summary statistics derived from sex-stratified tests and for the Z-score test, the SexDiffZscore() calculates the difference in any kind of measured entities such as effect size, SNP heritability, genetic correlation, etc., between sexes using a Z-score and associated p-value. This function also outputs the QQ plot of the Z-score distribution p-values. It is important to note that a study design that performs a sex-stratified analyses followed by a difference test equates to a SNP × Sex interaction test under certain conditions, specifically when there’s no interaction between covariates and sex, and when trait variances are similar across sexes 17 . This approach, while generally conservative, offers robustness against violations of these assumptions 17 , such as covariate interactions or sex differences in phenotypic variance. Another function, DiffZeroOne() , evaluates a Z-score to assess deviations from one and zero, testing the null hypothesis that a statistic (e.g., genetic correlation between males and females for a given trait) is less than one using a one-tailed test against a normal distribution. To apply these tests to a set of independent SNPs, GXwasR performs LD clumping using the LDclump() function Polygenic risk scores A polygenic risk score (PRS) is a quantitative estimate of an individual’s cumulative genetic predisposition to a specific trait. GXwasR provides several functions for working with PRSs in the context of sex differences: The ComputePRS() function generates polygenic risk scores, which are derived by summing the effects of multiple variants across the genome, each weighted by its effect size derived from GWAS. The function includes customizable Clumping + Thresholding (C + T) procedure to ensure that only independent trait-associated variants are used for PRS estimation. The function conducts an implicit model fitting process, wherein a number of p-value thresholds are tested to maximize the predictive power of the resulting polygenic scores. The function evaluates model performance using R 2 for quantitative traits and McFadden R 2 for binary traits 33 . Multiple plots are generated, including visualizations of the PRS model fit as a function of p-value thresholds, the distributions of computed PRS in males and females, and among cases and controls. For binary traits, ROC curves are produced (see Supplementary File). The ComputePRS() function, along with the SexRegress() function enable assessment of a genome-wide impact of sex differences on the genetic architecture of a trait which can manifest as sex differences in predictive performance of the PRS. For example, a PRS derived from one sex may predict trait status less accurately in the other sex 17 , 34 , 35 . In this context, SexRegress() facilitates the investigation of associations between a phenotype in the target sex using the PRS derived from effect sizes estimated in the other (discovery) sex. Differences may suggest a sex difference in the genetic burden of the trait, however, careful interpretation is warranted, as differences in sample size between sexes can impact prediction accuracy. Heritability and genetic correlation GXwasR performs two types of heritability estimation, (i) GREML: Genetic relatedness matrix (GRM) restricted maximum likelihood as implemented in GCTA 19 and (ii) LDSC: LD score regression 36 , 37 using the EstimateHerit() function. For the GREML model, EstimateHerit() fits the effects of all the SNPs as random effects in a mixed linear model. The function then estimates the variance explained by the SNPs using RELM (i.e., Restricted Maximum Likelihood) relying on the GRM. For ChrX, it adjusts for dosage compensation by modifying the GRM. For the LDSC model, EstimateHerit() uses GWAS summary statistics and recomputed LD scores 37 . EstimateHerit() also matches genetic variants across datasets by flipping and strand reversing and provides the option to use LD-clumped SNPs in the computation of LD scores. For both models, EstimateHerit() provides the trait heritability estimate in both observed and liability scales, and can partition heritability by MAF, chromosome, etc. It plots the estimated heritability vs. chromosome length, number of genes, and proportion of the SNPs per chromosome. The functions GeneticCorrBT() and SumstatGenCorr() estimate genetic correlation between two quantitative traits, two binary traits, and between a quantitative trait and a binary trait. GeneticCorrBT() performs a bivariate GREML 19 , 38 analysis utilizing genotype data and provides the option to compute genetic correlation using user defined MAF thresholds. SumstatGenCorr() applies High Definition Likelihood (HDL) 39 method to two sets of GWAS summary statistics, e.g., sex-stratified summary statistics. Both functions provide the option to compute the genetic correlation separately for each chromosome. Meta-analysis The MetaGwas() function combines summary statistics from separate GWASes. It utilizes an inverse variance-based approach to enable fixed-effect and random-effect models, and quantifies effect size heterogeneity using Cochran’s Q 40 and the I 2 statistics 41 . This function also calculates weighted Z-score-based p-values 42 . The function can also apply correction for genomic inflation 43 . The MetaGwas() function must be applied to independent trait-associated variants or those in linkage equilibrium. The ClumpLD() function and the distinct loci provided as the SNPfile input can be used for this purpose. The MetaGwas() function also generates Manhattan, QQ, and forest plots. Gene-based tests The TestXGene() function enables gene-based association testing, utilizing GWAS/XWAS summary statistics and SNP-SNP correlation matrices using either user-provided genotype data or 1000 Genomes Phase 3 reference data 44 . While TestXGene() is by default suited for X-linked gene-based tests, its design also allows for its application across the autosomes. The function provides flexibility to use pre-defined or custom genomic gene coordinates. Central to TestXGene() is its capability to compute gene-based SNP-SNP correlation matrices. The function is able to perform a variety of gene-based tests, including the burden test (BT 45 ), sequence kernel association test (SKAT 46 ), SKAT-O 47 (a combination of BT and SKAT), sum of χ 2 -statistics, aggregated Cauchy association test (ACAT 48 ), principal component approach (PCA 49 ), functional multiple linear regression model (FLM 50 ), Bonferroni correction test (simpleM 51 ), and minimum P-value (minp 51 ). These tests are executed by leveraging the functionality of PLINK1.9 and sumFREGAT 51 tools. Tutorials for the functions above as well as more detailed descriptions of the statistical models are provided in Supplementary File. Application to human height We illustrate the use of GXwasR to assess sex differences in the genetic basis of height using data from the Vanderbilt University Medical Center’s (VUMC) genetic biobank (BioVU) and the linked de-identified mirror of the VUMC electronic health record, known as the Synthetic Derivative (SD). Briefly, we used genotype data from 50,294 individuals (males = 21,881; females = 28,421) whose genetic data clustered with European reference populations and who had clinical data on height. After performing sex-aware genomic quality control using GXwasR, 3,359,461 variants were available for running sex-stratified GWAS using GXwas() function (see Supplemental File ). We identified 1,549 SNPs associated with height across both males and females (p < 5×10 −8 ) from 71 independent loci ( Supplemental File ) using ClumpLD() function ( Supplemental File ). These included five loci identified only in females on chromosomes 9 (rs4385527; rs2297086) and 15 (rs16869; rs4843154; rs4414460), and one locus identified only in males, on chromosome 2 (rs12615742). Fifteen of these loci were novel ( Table 2 ); for the other 56 loci, the lead or one of the clumped variants was previously reported to be associated with height in the NHGRI-EBI GWAS Catalog 52 . Loci associated in only one sex were more likely to be novel ( p-value from Fisher’s exact test = 0.006). One novel locus on ChrX was identified near pseudogene RP1-164L12.1/CORO1C pseudogene 1. View this table: View inline View popup Table 1: An overview of GXwasR’s functions ranging from quality control to statistical analysis. Note that these functions can be used across multiple categories. For example, QCsnp and QCsample can be used in both pre- and post-imputation. Detailed descriptions of these functions and example code are provided (see Supplementary File). View this table: View inline View popup Download powerpoint Table 2: Novel significant lead variants (p < 5×10 −8 ) from genome-wide association testing with height in BioVU data. While this locus is novel, other variants near RP1-164L12.1/CORO1C pseudogene 1 have previously been reported to be associated with height 53 , 54 . The t-test for sex differences was performed via SexDiff() on variants (see Supplementary File) with a nominal association (p < 0.05) in at least one sex. Thirty-four variants from four independent loci were identified as having different effects by sex (Benjamini-Hochberg adjusted p < 0.05). None of these variants were reported in the NHGRI-EBI GWAS Catalog as being associated with height. However, other variants mapping to three of the four annotated genes ( Table 3 ) have previously been reported to be associated with height 55 – 57 . View this table: View inline View popup Download powerpoint Table 3: Independent loci with sex-differentiated genetic effects on height (p < 0.05) after running the SexDiff() function with BioVU data. Heritability was estimated using the EstimateHerit() function, which implements both GREML and LDSC models. For GREML, the function computed the GRM from BioVU genotype data across chromosomes 1–23 and estimated chromosome-specific ( Figure 6 – 7 ) as well as genome-wide heritability in each sex (see Supplemental File). Download figure Open in new tab Figure 6. Male heritability per chromosome for height as estimated using EstimateHerit() function with BioVU data and GREML models. Download figure Open in new tab Figure 7. Female heritability per chromosome for height as estimated using EstimateHerit() function with BioVU data and GREML models. For LDSC, the function applied the LDSC model using sex-stratified GWAS and XWAS summary statistics, together with precomputed LD scores derived from UK Biobank genotype data 36 , to compute genome-wide heritability (see Supplemental File). Genome-wide height heritability estimates were similar between the two sexes and recapitulated previously reported heritability estimates ( Table 4 ). Male height heritability ranged between 0.21-0.34, and female height heritability between 0.24-0.36. The GREML model that utilized GRMs estimated a higher heritability compared to the LDSC model that used GWAS summary statistics and precomputed scores. View this table: View inline View popup Download powerpoint Table 4: Sex-stratified heritability estimates with associated standard error (SE) for height as generated via the EstimateHerit() using both GREML and LDSC models and BioVU data. We computed the genetic correlation between male and female BioVU summary statistics for height to be 0.748 (p = 5.10 × 10 −52 ), as shown in Table 5 , using the SumstatGenCorr() function. This analysis leveraged the function’s option to use a reference LD matrix computed from 1,029,876 imputed and quality-controlled SNPs from the UK Biobank dataset 39 (see Supplemental File). We also used SumstatGenCorr() to assess the genetic correlation between the BioVU sex-stratified summary statistics and publicly available sex-stratified summary statistics from the UK Biobank 58 for the raw standing height phenotype. These results are also presented in Table 5 . High genetic correlations were observed, consistent with previous findings 59 . View this table: View inline View popup Download powerpoint Table 5: Heritability and Genetic Correlation estimated using SumstatGenCorr() using BioVU and UKBB summary statistics of standing height. Discussion The exploration of sex-dependent genetic effects in the analysis of complex traits has gained momentum 60 , in part due to increasing awareness of sex differences in health and disease, but also due to recognition of the need to study diverse cohorts in biomedical research to reduce health disparities. As previously described 17 , a variety of approaches can be employed to assess different aspects of sex differences in genetic effects. We recommend the use of multiple approaches, particularly when the specific nature of allelic effects is unknown. The chosen approach might vary depending on the allelic effects, the size of the sample, and specific characteristics of the dataset or trait being studied. To facilitate these analyses, we have created GXwasR, a comprehensive R package designed to enable these complex analyses. It enables sex-aware and sex-agnostic GWAS and XWAS, equipped with functionalities that span the entire analytical spectrum, from preliminary QC to association analysis strategies. It facilitates sex-combined and sex-stratified analysis methods across the genome, managing XCI intricacies. Importantly, it can integrate with other workflows at multiple stages. To demonstrate the capabilities of GXwasR, we provided several tutorials using simulated datasets, along with an applied analysis of height GWAS data from our BioVU, the VUMC biobank, showcasing its utility in investigating sex-specific genetic architecture of a complex trait. Our ultimate goal with GXwasR is to increase accessibility of these types of analyses to a broad scientific audience. GXwasR is developed as an R package. To install the development version, use the command pak::pak(“boseb/GXwasR”). The source code is publicly available on GitHub at https://github.com/boseb/GXwasRunder the Creative Commons Attribution-NonCommercial 4.0 International Public License. The scripts for running all functions and accompanying tutorials are provided in the Supplementary File. Data Availability The data that support the findings of this study are available from Vanderbilt University Medical Center but restrictions apply to the availability of these data, which were used under license for the current study, and so are not publicly available. Data are, however, available from the authors upon reasonable request and with permission of Vanderbilt University Medical Center. The data in question must first be reviewed by the Integrated Data Access and Services Core to ensure that the de-identification is complete and no potentially identifying information remains. Please contact the Vanderbilt Institute for Clinical and Translational Research ( research.support.services{at}vumc.org ) for more information. Acknowledgements We would like to thank Ekaterina Khramtsova and Catherine Stanhope for helpful comments during the development of the package. This research was supported in part through the computational resources and staff contributions provided by the Genomics Compute Cluster which is jointly supported by the Feinberg School of Medicine, the Center for Genetic Medicine, and Feinberg’s Department of Biochemistry and Molecular Genetics, the Office of the Provost, the Office for Research, and Northwestern Information Technology. The Genomics Compute Cluster is part of Quest, Northwestern University’s high performance computing facility, with the purpose of advancing research in genomics. Footnotes ↵ * co-last author References 1. ↵ Clocchiatti , A. , Cora , E. , Zhang , Y. & Dotto , G. P. Sexual dimorphism in cancer . Nat Rev Cancer 16 , 330 – 339 ( 2016 ). OpenUrl CrossRef PubMed 2. ↵ Zheng , D. et al. Sexual dimorphism in the incidence of human cancers . BMC Cancer 19 , 684 ( 2019 ). OpenUrl CrossRef PubMed 3. ↵ Franconi , F. & Campesi , I. Sex and gender influences on pharmacological response: an overview . Expert Rev Clin Pharmacol 7 , 469 – 485 ( 2014 ). OpenUrl CrossRef PubMed 4. ↵ Heidari , S. , Babor , T. F. , De Castro , P. , Tort , S. & Curno , M. Sex and Gender Equity in Research: rationale for the SAGER guidelines and recommended use . Research Integrity and Peer Review 1 , 2 ( 2016 ). OpenUrl 5. ↵ Ober , C. , Loisel , D. A. & Gilad , Y. Sex-Specific Genetic Architecture of Human Disease . Nat Rev Genet 9 , 911 – 922 ( 2008 ). OpenUrl CrossRef PubMed Web of Science 6. Parsch , J. & Ellegren , H. The evolutionary causes and consequences of sex-biased gene expression . Nat Rev Genet 14 , 83 – 87 ( 2013 ). OpenUrl CrossRef PubMed 7. ↵ Zuo , L. et al. Sex difference of autosomal alleles in populations of European and African descent . Genes Genomics 37 , 1007 – 1016 ( 2015 ). OpenUrl PubMed 8. ↵ Tukiainen , T. et al. Landscape of X chromosome inactivation across human tissues . Nature 550 , 244 – 248 ( 2017 ). OpenUrl CrossRef PubMed 9. ↵ Wise , A. L. , Gyi , L. & Manolio , T. A. eXclusion: Toward Integrating the X Chromosome in Genome-wide Association Analyses . Am J Hum Genet 92 , 643 – 647 ( 2013 ). OpenUrl CrossRef PubMed 10. ↵ Spatz , A. , Borg , C. & Feunteun , J. X-Chromosome Genetics and Human Cancer . Nat Rev Cancer 4 , 617 – 629 ( 2004 ). OpenUrl CrossRef PubMed Web of Science 11. Arnold , A. P. et al. The importance of having two X chromosomes . Philos Trans R Soc Lond B Biol Sci 371 , 20150113 ( 2016 ). OpenUrl CrossRef PubMed 12. Martin , J. et al. Examining Sex-Differentiated Genetic Effects Across Neuropsychiatric and Behavioral Traits . Biological Psychiatry 89 , 1127 – 1137 ( 2021 ). OpenUrl CrossRef PubMed 13. Lahn , B. T. & Page , D. C. Functional Coherence of the Human Y Chromosome . Science 278 , 675 – 680 ( 1997 ). OpenUrl Abstract / FREE Full Text 14. ↵ Sun , L. , Wang , Z. , Lu , T. , Manolio , T. A. & Paterson , A. D. eXclusionarY: 10 years later, where are the sex chromosomes in GWASs? Am J Hum Genet 110 , 903 – 912 ( 2023 ). OpenUrl CrossRef PubMed 15. ↵ Chang , D. et al. Accounting for eXentricities: Analysis of the X Chromosome in GWAS Reveals X-Linked Genes Implicated in Autoimmune Diseases . PLOS ONE 9 , e113684 ( 2014 ). OpenUrl CrossRef PubMed 16. ↵ Gorlov , I. P. & Amos , C. I. Why does the X chromosome lag behind autosomes in GWAS findings? PLOS Genetics 19 , e1010472 ( 2023 ). OpenUrl 17. ↵ Khramtsova , E. A. et al. Quality control and analytic best practices for testing genetic models of sex differences in large populations . Cell 186 , 2044 – 2061 ( 2023 ). OpenUrl CrossRef PubMed 18. ↵ Purcell , S. et al. PLINK: A Tool Set for Whole-Genome Association and Population-Based Linkage Analyses . American Journal of Human Genetics 81 , 559 ( 2007 ). OpenUrl CrossRef PubMed 19. ↵ Yang , J. , Lee , S. H. , Goddard , M. E. & Visscher , P. M. GCTA: A Tool for Genome-wide Complex Trait Analysis . Am J Hum Genet 88 , 76 – 82 ( 2011 ). OpenUrl CrossRef PubMed 20. ↵ Mbatchou , J. et al. Computationally efficient whole-genome regression for quantitative and binary traits . Nat Genet 53 , 1097 – 1103 ( 2021 ). OpenUrl CrossRef PubMed 21. ↵ Zhou , W. et al. Efficiently controlling for case-control imbalance and sample relatedness in large-scale genetic association studies . Nat Genet 50 , 1335 – 1341 ( 2018 ). OpenUrl CrossRef PubMed 22. ↵ Marees , A. T. et al. A tutorial on conducting genome-wide association studies: Quality control and statistical analysis . Int J Methods Psychiatr Res 27 , e1608 ( 2018 ). OpenUrl CrossRef PubMed 23. ↵ Turner , S. et al. Quality Control Procedures for Genome Wide Association Studies . Curr Protoc Hum Genet CHAPTER , Unit1.19 ( 2011 ). 24. ↵ Flaquer , A. , Rappold , G. A. , Wienker , T. F. & Fischer , C. The human pseudoautosomal regions: a review for genetic epidemiologists . Eur J Hum Genet 16 , 771 – 779 ( 2008 ). OpenUrl CrossRef PubMed Web of Science 25. ↵ Cotter , D. J. , Brotman , S. M. & Wilson Sayres , M. A. Genetic Diversity on the Human X Chromosome Does Not Support a Strict Pseudoautosomal Boundary . Genetics 203 , 485 – 492 ( 2016 ). OpenUrl Abstract / FREE Full Text 26. ↵ Bhowmick , B. K. , Satta , Y. & Takahata , N. The origin and evolution of human ampliconic gene families and ampliconic structure . Genome Res 17 , 441 – 450 ( 2007 ). OpenUrl Abstract / FREE Full Text 27. ↵ Disteche , C. M. Dosage compensation of the sex chromosomes and autosomes . Seminars in Cell & Developmental Biology 56 , 9 – 18 ( 2016 ). OpenUrl CrossRef PubMed 28. ↵ Fang , H. , Disteche , C. M. & Berletch , J. B. X Inactivation and Escape: Epigenetic and Structural Features . Frontiers in Cell and Developmental Biology 7 , ( 2019 ). 29. ↵ Kotz , S. & Johnson , N.L. Fisher , R. A. Statistical Methods for Research Workers . in Breakthroughs in Statistics: Methodology and Distribution (eds. Kotz , S. & Johnson , N.L. ) 66 – 70 ( Springer , New York, NY , 1992 ). doi: 10.1007/978-1-4612-4380-9_6 . OpenUrl CrossRef 30. ↵ Rhodes , D. R. , Barrette , T. R. , Rubin , M. A. , Ghosh , D. & Chinnaiyan , A. M. Meta-analysis of microarrays: interstudy validation of gene expression profiles reveals pathway dysregulation in prostate cancer . Cancer Res 62 , 4427 – 4433 ( 2002 ). OpenUrl Abstract / FREE Full Text 31. ↵ Stouffer , S.A. , Suchman , E.A. , DeVinney , L.C. , Star , S.A. and Williams Jr , R.M. , 1949 . The american soldier: Adjustment during army life.(studies in social psychology in world war ii) , vol. 1 . 32. ↵ Su , Y. et al. XCMAX4: A Robust X Chromosomal Genetic Association Test Accounting for Covariates . Genes (Basel) 13 , 847 ( 2022 ). OpenUrl PubMed 33. ↵ Hardin , J. W. & Hilbe , J. M. Generalized Linear Models and Extensions, Second Edition . ( Taylor & Francis , 2007 ). 34. ↵ Koch , E. et al. Sex-specific effects of polygenic risk for schizophrenia on lifespan cognitive functioning in healthy individuals . Transl Psychiatry 11 , 520 ( 2021 ). OpenUrl PubMed 35. ↵ Zhang , C. , Ye , Y. & Zhao , H. Comparison of Methods Utilizing Sex-Specific PRSs Derived From GWAS Summary Statistics . Front Genet 13 , 892950 ( 2022 ). OpenUrl CrossRef PubMed 36. ↵ Bulik-Sullivan , B. K. et al. LD Score regression distinguishes confounding from polygenicity in genome-wide association studies . Nat Genet 47 , 291 – 295 ( 2015 ). OpenUrl CrossRef PubMed 37. ↵ Privé , F. , Arbel , J. & Vilhjálmsson , B. J. LDpred2: better, faster, stronger . Bioinformatics 36 , 5424 – 5431 ( 2021 ). OpenUrl CrossRef PubMed 38. ↵ Lee , S. H. , Yang , J. , Goddard , M. E. , Visscher , P. M. & Wray , N. R. Estimation of pleiotropy between complex diseases using single-nucleotide polymorphism-derived genomic relationships and restricted maximum likelihood . Bioinformatics 28 , 2540 – 2542 ( 2012 ). OpenUrl CrossRef PubMed Web of Science 39. ↵ Ning , Z. , Pawitan , Y. & Shen , X. High-definition likelihood inference of genetic correlations across human complex traits . Nat Genet 52 , 859 – 864 ( 2020 ). OpenUrl CrossRef PubMed 40. ↵ Cochran , W. G. THE COMPARISON OF PERCENTAGES IN MATCHED SAMPLES . Biometrika 37 , 256 – 266 ( 1950 ). OpenUrl CrossRef PubMed Web of Science 41. ↵ Huedo-Medina , T. B. , Sánchez-Meca , J. , Marín-Martínez , F. & Botella , J. Assessing heterogeneity in meta-analysis: Q statistic or I2 index? Psychol Methods 11 , 193 – 206 ( 2006 ). OpenUrl CrossRef PubMed Web of Science 42. ↵ Willer , C. J. , Li , Y. & Abecasis , G. R. METAL: fast and efficient meta-analysis of genomewide association scans . Bioinformatics 26 , 2190 ( 2010 ). OpenUrl CrossRef PubMed Web of Science 43. ↵ Devlin , B. , Roeder , K. & Wasserman , L. Genomic control, a new approach to genetic-based association studies . Theor Popul Biol 60 , 155 – 166 ( 2001 ). OpenUrl CrossRef PubMed Web of Science 44. ↵ Auton , A. et al. A global reference for human genetic variation . Nature 526 , 68 – 74 ( 2015 ). OpenUrl CrossRef PubMed 45. ↵ Dering , C. , Hemmelmann , C. , Pugh , E. & Ziegler , A. Statistical analysis of rare sequence variants: an overview of collapsing methods . Genetic Epidemiology 35 , S12 – S17 ( 2011 ). OpenUrl CrossRef PubMed 46. ↵ Chen , H. , Meigs , J. B. & Dupuis , J. Sequence kernel association test for quantitative traits in family samples . Genet Epidemiol 37 , 196 – 204 ( 2013 ). OpenUrl CrossRef PubMed 47. ↵ Lee , S. et al. Optimal unified approach for rare-variant association testing with application to small-sample case-control whole-exome sequencing studies . Am J Hum Genet 91 , 224 – 237 ( 2012 ). OpenUrl CrossRef PubMed 48. ↵ Liu , Y. et al. ACAT: A Fast and Powerful p Value Combination Method for Rare-Variant Analysis in Sequencing Studies . Am J Hum Genet 104 , 410 – 421 ( 2019 ). OpenUrl CrossRef PubMed 49. ↵ Wang , K. & Abbott , D. A principal components regression approach to multilocus genetic association studies . Genet Epidemiol 32 , 108 – 118 ( 2008 ). OpenUrl CrossRef PubMed Web of Science 50. ↵ Fan , R. et al. Functional linear models for association analysis of quantitative traits . Genet Epidemiol 37 , 726 – 742 ( 2013 ). OpenUrl CrossRef PubMed 51. ↵ Svishcheva , G. R. , Belonogova , N. M. , Zorkoltseva , I. V. , Kirichenko , A. V. & Axenovich , T. I. Gene-based association tests using GWAS summary statistics . Bioinformatics 35 , 3701 – 3708 ( 2019 ). OpenUrl CrossRef PubMed 52. ↵ Cerezo , M. et al. The NHGRI-EBI GWAS Catalog: standards for reusability, sustainability and diversity . Nucleic Acids Research 53 , D998 – D1005 ( 2025 ). OpenUrl CrossRef PubMed 53. ↵ Sakaue , S. et al. A cross-population atlas of genetic associations for 220 human phenotypes . Nat Genet 53 , 1415 – 1424 ( 2021 ). OpenUrl CrossRef PubMed 54. ↵ Gudbjartsson , D. F. et al. Many sequence variants affecting diversity of adult human height . Nat Genet 40 , 609 – 615 ( 2008 ). OpenUrl CrossRef PubMed Web of Science 55. ↵ Kichaev , G. et al. Leveraging Polygenic Functional Enrichment to Improve GWAS Power . Am J Hum Genet 104 , 65 – 75 ( 2019 ). OpenUrl CrossRef PubMed 56. Yengo , L. et al. A saturated map of common genetic variants associated with human height . Nature 610 , 704 – 712 ( 2022 ). OpenUrl CrossRef PubMed 57. ↵ Richardson , T. G. , Sanderson , E. , Elsworth , B. , Tilling , K. & Davey Smith , G. Use of genetic variation to separate the effects of early and later life adiposity on disease risk: mendelian randomisation study . BMJ 369 , m1203 ( 2020 ). OpenUrl Abstract / FREE Full Text 58. ↵ http://www.nealelab.is/uk-biobank/ 59. ↵ Bernabeu , E. et al. Sex differences in genetic architecture in the UK Biobank . Nat Genet 53 , 1283 – 1289 ( 2021 ). OpenUrl CrossRef PubMed 60. ↵ Khramtsova , E. A. , Davis , L. K. & Stranger , B. E. The role of sex in the genomics of human complex traits . Nat Rev Genet 20 , 173 – 190 ( 2019 ). OpenUrl CrossRef PubMed View the discussion thread. Back to top Previous Next Posted June 11, 2025. Download PDF Supplementary Material Data/Code Email Thank you for your interest in spreading the word about medRxiv. NOTE: Your email address is requested solely to identify you as the sender of this article. Your Email * Your Name * Send To * Enter multiple addresses on separate lines or separate them with commas. You are going to email the following GXwasR: A Toolkit for Investigating Sex-Differentiated Genetic Effects on Complex Traits Message Subject (Your Name) has forwarded a page to you from medRxiv Message Body (Your Name) thought you would like to see this page from the medRxiv website. Your Personal Message CAPTCHA This question is for testing whether or not you are a human visitor and to prevent automated spam submissions. Share GXwasR: A Toolkit for Investigating Sex-Differentiated Genetic Effects on Complex Traits Banabithi Bose , Freida Blostein , Jeewoo Kim , Jessica Winters , Ky’Era V. Actkins , David Mayer , Harrsha Congivaram , Maria Niarchou , Digna Velez Edwards , Lea K. Davis , Barbara E. Stranger medRxiv 2025.06.10.25329327; doi: https://doi.org/10.1101/2025.06.10.25329327 Share This Article: Copy Citation Tools GXwasR: A Toolkit for Investigating Sex-Differentiated Genetic Effects on Complex Traits Banabithi Bose , Freida Blostein , Jeewoo Kim , Jessica Winters , Ky’Era V. Actkins , David Mayer , Harrsha Congivaram , Maria Niarchou , Digna Velez Edwards , Lea K. Davis , Barbara E. Stranger medRxiv 2025.06.10.25329327; doi: https://doi.org/10.1101/2025.06.10.25329327 Citation Manager Formats BibTeX Bookends EasyBib EndNote (tagged) EndNote 8 (xml) Medlars Mendeley Papers RefWorks Tagged Ref Manager RIS Zotero Tweet Widget Facebook Like Google Plus One Subject Area Genetic and Genomic Medicine Subject Areas All Articles Addiction Medicine (568) Allergy and Immunology (863) Anesthesia (299) Cardiovascular Medicine (4425) Dentistry and Oral Medicine (443) Dermatology (382) Emergency Medicine (607) Endocrinology (including Diabetes Mellitus and Metabolic Disease) (1507) Epidemiology (15221) Forensic Medicine (30) Gastroenterology (1123) Genetic and Genomic Medicine (6588) Geriatric Medicine (667) Health Economics (997) Health Informatics (4524) Health Policy (1368) Health Systems and Quality Improvement (1612) Hematology (540) HIV/AIDS (1264) Infectious Diseases (except HIV/AIDS) (15910) Intensive Care and Critical Care Medicine (1103) Medical Education (623) Medical Ethics (145) Nephrology (667) Neurology (6588) Nursing (346) Nutrition (998) Obstetrics and Gynecology (1143) Occupational and Environmental Health (956) Oncology (3331) Ophthalmology (970) Orthopedics (369) Otolaryngology (420) Pain Medicine (435) Palliative Medicine (129) Pathology (663) Pediatrics (1690) Pharmacology and Therapeutics (691) Primary Care Research (710) Psychiatry and Clinical Psychology (5440) Public and Global Health (9219) Radiology and Imaging (2195) Rehabilitation Medicine and Physical Therapy (1369) Respiratory Medicine (1196) Rheumatology (593) Sexual and Reproductive Health (710) Sports Medicine (529) Surgery (710) Toxicology (99) Transplantation (289) Urology (265) (function(){function c(){var b=a.contentDocument||a.contentWindow.document;if(b){var d=b.createElement('script');d.innerHTML="window.__CF$cv$params={r:'9ffcc6f63d5ddf88',t:'MTc3OTQ2MjkzNg=='};var a=document.createElement('script');a.src='/cdn-cgi/challenge-platform/scripts/jsd/main.js';document.getElementsByTagName('head')[0].appendChild(a);";b.getElementsByTagName('head')[0].appendChild(d)}}if(document.body){var a=document.createElement('iframe');a.height=1;a.width=1;a.style.position='absolute';a.style.top=0;a.style.left=0;a.style.border='none';a.style.visibility='hidden';document.body.appendChild(a);if('loading'!==document.readyState)c();else if(window.addEventListener)document.addEventListener('DOMContentLoaded',c);else{var e=document.onreadystatechange||function(){};document.onreadystatechange=function(b){e(b);'loading'!==document.readyState&&(document.onreadystatechange=e,c())}}}})();