A multi-trait genome-wide association study of coronary artery disease and subclinical atherosclerosis traits

A multi-trait genome-wide association study of coronary artery disease and subclinical atherosclerosis traits | Research Square window.SnipcartSettings = { analytics: { enabled: false } }; (function() { var accessVector = localStorage.getItem('access_vector') || ''; window.dataLayer = window.dataLayer || []; if (accessVector) { window.dataLayer.push({ user: { profile: { profileInfo: { snid: accessVector } } } }); } })(); (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],j=d.createElement(s),dl=l!='dataLayer'?'&l='+l:'';j.async=true;j.src='https://www.googletagmanager.com/gtm.js?id='+i+dl;f.parentNode.insertBefore(j,f);})(window,document,'script','dataLayer','GTM-K279D39R'); Browse Preprints In Review Journals COVID-19 Preprints AJE Video Bytes Research Tools Research Promotion AJE Professional Editing AJE Rubriq About Preprint Platform In Review Editorial Policies Our Team Advisory Board Help Center Sign In Submit a Preprint Cite Share Download PDF Article A multi-trait genome-wide association study of coronary artery disease and subclinical atherosclerosis traits Natalie R. Hasbani, Adam S. Heath, Chani J. Hodonsky, Julie Hahn, and 11 more This is a preprint; it has not been peer reviewed by a journal. https://doi.org/ 10.21203/rs.3.rs-6456056/v1 This work is licensed under a CC BY 4.0 License Status: Posted Version 1 posted You are reading this latest preprint version Abstract Measures of subclinical atherosclerosis, such as coronary artery calcification (CAC) and carotid intima-media thickness (CIMT), reflect the underlying pathophysiology of coronary artery disease (CAD) and are genetically correlated with CAD and related risk factors. Leveraging summary statistics from genome-wide association studies of CAD, CIMT, CAC, type 2 diabetes, low-density lipoprotein cholesterol, and systolic blood pressure, we performed 15 separate multi-trait GWAS to identify shared susceptibility loci and elucidate the pleiotropic architecture underlying atherosclerosis. We identified 442 shared risk loci across all analyses that met an experiment-wide Bonferroni threshold of 3.3 × 10 -9 , uncovering 195 novel atherosclerosis loci. Multi-trait colocalization confirmed a shared causal signal in 25 shared novel loci for atherosclerosis. Trait-eQTL colocalization identified evidence of a shared causal signal in arterial, subcutaneous adipose, and cardiac tissues, implicating genes such as PRRX2 , BNC2 , CLIC4 , SCAI , and PPP6C , and pathways related to vascular remodeling, inflammation, and metabolic regulation. Health sciences/Cardiology/Cardiovascular biology/Cardiovascular genetics Health sciences/Cardiology/Cardiovascular biology/Cardiovascular diseases/Vascular diseases/Atherosclerosis Figures Figure 1 Figure 2 Figure 3 Figure 4 Figure 5 Figure 6 Figure 7 Introduction Coronary artery disease (CAD) is a complex disease process involving a convergence of environmental and genetic risk factors. Several underlying heritable traits exist which indicate the presence of atherosclerosis prior to the clinical manifestation of CAD, including coronary artery calcification (CAC) and carotid intima- media thickness (CIMT). Recent genetic studies also provide abundant evidence of pleiotropy among risk factors with approximately 50% of identified CAD risk loci associating with underlying clinical risk factors. 1-11 Genome-wide association studies (GWAS) for direct and indirect measures of atherosclerosis as well as associated risk factors such as systolic blood pressure (SBP), low-density lipoprotein cholesterol (LDL-C), and type 2 diabetes (T2D) have identified hundreds of susceptibility loci. 10-20 However, the discovered loci collectively only explain a fraction of the heritability of these phenotypes, suggesting that additional associated susceptibility loci remain to be discovered. 21 Combining information from the shared genetic architecture between atherosclerosis measures and clinical risk factors may help identify new susceptibility loci with shared underlying biology. Statistical approaches that facilitate the joint analysis of multiple correlated traits have been developed that increase the power to detect loci that are pleiotropically associated with more than one trait. 22,23 These methods have thus far uncovered many shared susceptibility loci for cardiometabolic phenotypes and provided unique insight into shared genetic pathways. 24-26 No studies to date have used multi-trait approaches to evaluate both subclinical and clinical atherosclerosis phenotypes along with risk factors. Joint analysis of clinical CAD with subclinical atherosclerosis traits and related risk factors may uncover key susceptibility loci and pleiotropic mechanisms contributing to the development of CAD. Here, we report a multi-stage, multi-trait GWAS of atherosclerosis-related phenotypes and selected cardiovascular risk factors designed to provide deeper insights into the shared genetic architecture underlying atherosclerosis and enhance power for locus discovery. Results Cross-trait analysis for atherosclerosis and related risk factors We conducted a multi-stage, multi-trait analysis using GWAS summary statistics collected from 3 different measures of atherosclerosis (CAD, CAC, and CIMT) and 3 risk factors (T2D, LDL-C, SBP, Fig. 1 , Supplemental Table 1 ). 10 , 12 , 14 , 16 , 17 , 19 , 20 First, we conducted cross-trait linkage disequilibrium (LD) score regression using LD Score v1.0 (LDSC) to estimate genetic correlation, and shared heritability for each of the 12 pairs of traits. 27 Multi-ancestry GWAS summary statistics were available for all traits except SBP where summary statistics were only available from those of European ancestry at the time of this analysis. Analyses were performed using the 1000 Genomes European LD reference panel. 28 All atherosclerosis and related risk factor traits were significantly (P < 0.05) genetically correlated with CAD (Fig. 2 ). Both CAC and CIMT were significantly genetically correlated with T2D and SBP. CAC was also significantly genetically correlated with LDL-C, but CIMT was not. The strongest observed genetic correlation across all traits was between CAC and CAD [ r g =0.74 ± 0.05 (standard error)], followed by CAD and T2D [ r g =0.34 ± 0.02]. For studies that had ancestry-specific datasets available, we repeated the analysis using European-only datasets and found the genetic correlations to be similar with only slight variations in the magnitude and precision of the estimates ( Supplemental Table 2 , Fig. 2 ). ​ Multi-trait GWAS We conducted a multi-stage, multi-trait GWAS using N-weighted multivariate genome-wide association meta-analysis (N-GWAMA). 23 N-GWAMA applies the estimates from cross-trait LD-score regression to re-weight test statistics from single-trait GWAS summary statistics, by sample size and estimated heritability, while adjusting for genetic covariance across traits. We conducted 15 multi-trait analyses (Fig. 1 ) across three stages to systematically assess the genetic architecture of atherosclerosis-related traits and risk factors. Stage 1 focused on atherosclerosis traits (CAD, CAC, CIMT) to capture shared genetic associations within clinically relevant disease endpoints. Stage 2 incorporated subclinical traits and selected risk factors (e.g., LDL-C, SBP, T2D) to assess their independent and joint contributions. Stage 3 integrated both approaches, combining atherosclerosis traits with risk factors (e.g., CAD-CAC-SBP) to dissect the genetic pathways underlying disease progression from subclinical to clinical disease. By structuring our analysis in this manner, we aimed to maximize power for variant discovery while mitigating dilution effects from unassociated phenotypes. This approach also enabled us to differentiate variants associated with distinct biological pathways contributing to atherosclerosis development. We used Functional Mapping and Annotation of GWAS (FUMA) to annotate the summary results generated from N-GWAMA. 29 We identified 1,177 multi-trait risk loci at a GWAS significance threshold of 5×10 − 8 , of which 948 met the experiment-wide Bonferroni-corrected significance threshold of 3.3×10 − 9 (5×10 − 8 ÷15) across all stages of analysis ( Supplemental Fig. 1–3 ). A risk locus was considered novel for atherosclerosis if the lead variant did not meet the genome-wide significance threshold in any of the initial single-trait summary statistics for CAC, CAD, or CIMT and was located more than 500 kb away from a previously reported genome-wide significant variant in the GWAS Catalog. 30 For Stage 2 analyses which were restricted to subclinical atherosclerosis traits, a locus was considered novel only if it had not been reported in the initial CAD summary statistics and was also more than 500 kb away from any genome-wide significant variant for CAD. This approach ensured that all reported loci were novel for all measures of atherosclerosis regardless of analysis stage. We identified 173 significant loci during Stage 1 analyses. Most of the significant loci had known associations with at least one atherosclerosis trait (CAD, CAC, or CIMT) ( Supplemental Table 3 ). Only one locus, rs472784 in DLG2 (P = 2.0×10 − 9 ), from the CAD-CIMT analysis was novel for atherosclerosis. Similarly, in Stage 2 analyses, 451 experiment-wide significant loci were identified and 210 were novel atherosclerosis loci ( Supplemental Table 3, Supplemental Table 4 ). Of these, 10 loci were also novel for the included risk factor. Finally, during Stage 3 analyses, we identified 324 significant loci with 115 novel loci for atherosclerosis. There were 17 significant loci that were also novel for the selected risk factor in the analysis ( Supplemental Table 3, Supplemental Table 5 ). To identify shared and unique loci across all multi-trait analyses, we merged characterized loci into a single shared genomic locus if the lead variants from different analyses were within 500 kb of each other. For each shared locus, the variant with the most significant P-value across analyses was designated as the shared lead variant. Thus, the 1,177 multi-trait risk loci at GWAS significance threshold of 5×10 − 8 collapsed into 535 shared loci across all stages of analysis, with 442 containing at least one experiment-wide significant multi-trait locus (Fig. 3 , Supplemental Table 6) . Most of the 442 significant shared risk loci were known genomic regions that have previously been associated with atherosclerosis, with 195 shared risk loci that were novel for atherosclerosis. There were 25 loci that were also novel for a selected risk factor in the multi-trait analysis. Half of the novel atherosclerosis loci were identified in analyses restricted to subclinical atherosclerosis (101/195 = 51%). Overall, there were 60 novel atherosclerosis loci associated with CIMT, 27 loci associated with CAC, and 14 novel atherosclerosis loci overlapping with CAC and CIMT. The remaining novel atherosclerosis loci were identified in analyses with CAD (N = 94). There were 41 novel atherosclerosis loci shared between CAD and CIMT, 30 shared between CAC and CAD and 23 shared across all atherosclerosis traits (CAD, CAC, and CIMT). For novel atherosclerosis loci, 4 regions overlapped the most frequently across all stages of multi-trait analyses (nearest gene: BNC2 , GPATCH2 , INSR , JAZF1 ). Most of the experiment-wide significant loci were identified in multi-trait analyses which included SBP (Fig. 3 ). Distinct groups of shared genomic regions were identified across atherosclerosis traits. There were 50 shared loci shared across SBP, CIMT, and CAD, 58 shared loci that included just CIMT-SBP, 37 shared across all atherosclerosis traits and SBP and 35 shared loci that included CAC-CAD-SBP. Similar patterns were noted with both remaining risk factor traits with various trait combinations identifying important pleiotropic genomic regions for atherosclerosis. Six shared loci that were shared across all traits in the analysis, all with known associations with atherosclerosis (nearest genes: MAT2A, IRS1, STAG1, PVRL2, OPRL1, ARVCF ). Trait-trait and Trait- eQTL colocalization Multi-trait colocalization was conducted using HyPrColoc, a Bayesian divisive clustering algorithm designed to identify shared causal signals within a genomic region. 22 . Evidence for colocalization was determined based on the default variant-specific regional and alignment priors (P R ∗=P A ∗ =0.5), with colocalization identified when P R P A ≥0.25. Strong evidence of colocalization across traits was defined as P R P A ≥ 0.80. Overall, 164 significant shared risk loci identified in the multi-trait analysis also had evidence of multi-trait colocalization with a measure of atherosclerosis in HyPrColoc. Overall, multi-trait GWAS analyses that included SBP also colocalized the most frequently with the respective atherosclerosis traits (N = 164), followed by LDL-C (N = 93), and T2D (N = 36). There were 25 novel atherosclerosis loci with evidence of colocalization (Supplemental Table 7). Of the 25 shared novel atherosclerosis loci with evidence of colocalization, 7 analyses strongly colocalized with P R P A ≥0.80 (nearest gene: BNC2, SCAI, TSC22D2, SRRM1, ABCB11, PRRX2 [Figure 4 ]). Finally, for loci that colocalized in the trait-trait analysis, we performed trait- expression quantitative loci (eQTL) colocalization using eQTL data from GTEX v8 using the COLOC package in R. 31 ,32 33 Using summary statistics from the multi-trait GWAS for corresponding trait-trait pairwise colocalization, we focused our analysis on a subset of GTEx v8 tissues selected based on their biological relevance to the studied traits ( Supplemental Table 8 ). Evidence for colocalization was defined as a posterior probability of a shared causal variant (PP H4 ) ≥ 0.50 and a conditional posterior probability (PP c =PP H4 /(PP H3 +PP H4) ) ≥ 0.80. Strong evidence for colocalization was defined as (PP H4 ) ≥ 0.80 and a conditional posterior probability (PP H4 /(PP H3 +PP H4 )) ≥ 0.80. We found evidence of colocalization (PP H4 ≥0.50 and PP c ≥0.80) with eQTL data from GTEx with novel atherosclerosis loci. Evidence of colocalization was identified most frequently with eQTLs in adipose tissue and arterial tissue of the Aorta and Tibia (N = 132, 114, and 112, respectively). There were 18 significant novel atherosclerosis loci with evidence of colocalization with eQTL data from GTEx that had evidence of trait-trait colocalization as well (Table 1 ). We identified 5 genes with strong evidence for colocalization (P R P A ≥0.80, PP H4 ≥0.80, and PP c ≥0.80) in trait-eQTL and in trait-trait colocalization analysis ( PRRX2 , BNC2 , CLIC4 , SCAI , and PPP6C). SBP and CIMT colocalized with expression of PRRX2, SCAI , and PPP6C in arterial tissues (Fig. 5 ). These findings suggest a potential mechanistic link between vascular gene regulation and atherosclerosis traits. The expression of BNC2 in whole blood colocalized with CAC, CAD, CIMT, and LDL-C (Fig. 6 ), while the expression of CLIC4 in visceral omentum adipose tissue and left ventricle heart tissue colocalized with CAD, CAC, CIMT, and SBP (Fig. 7 ). Table 1 Novel atherosclerosis risk loci that colocalized with tissue-specific gene expression levels. Shared Locus ID 1 Traits in Multi-trait GWAS Gene Symbol 2 Tissue Type 3 Posterior Probability (H4) 4 Conditional Posterior Probability 5 Candidate Causal SNP 276 CAC, CAD, LDL-C BNC2 Whole Blood 0.98 0.98 rs28498684 276 CAD, CIMT, LDL-C BNC2 Whole Blood 0.98 0.98 rs28498684 6 SBP, CAC, CAD CLIC4 Left Ventricle 0.93 0.94 rs4366267 Visceral Adipose 0.96 0.97 rs4366267 6 SBP, CIMT, CAD CLIC4 Left Ventricle 0.92 0.94 rs72654647 Visceral Adipose 0.97 0.97 rs6686889 287 SBP, CIMT PPP6C Tibial Artery 0.83 0.85 rs72765265 SCAI Aorta Artery 0.95 0.95 rs11793512 Tibial Artery 0.93 0.93 rs11793475 289** SBP, CIMT PRRX2 Aorta Artery 0.98 0.98 rs11788582 Tibial Artery 0.98 0.98 rs11788582 Left Ventricle 0.97 0.98 rs920659 Subcutaneous Adipose 0.80 0.84 rs920659 Visceral Omentum Adipose 0.98 0.98 rs13299355 Cultured fibroblasts Cells 0.98 0.98 rs59878076 Abbreviations: CAC, coronary artery calcification; CAD, coronary artery disease; CIMT. carotid intima- media thickness; GWAS, genome-wide association study; LDL-C, low- density lipoprotein cholesterol; SBP, systolic blood pressure, SNP, single nucleotide polymorphism 1 Shared Locus ID is a unique identifier for loci shared across multiple traits, defined as SNPs within 500 kb of each other. ** indicates novel for both atherosclerosis and associated risk factors. 2 HGNC Gene Symbol represents the gene whose eQTL colocalized with the genomic locus, suggesting a regulatory role. 3 Tissue indicates where the colocalized signal was detected, based on GTEx or other tissue-specific datasets. 4 Posterior Probability (H4) represents the probability that the signal is shared between traits under the H4 model of colocalization. 5 Conditional Posterior Probability reflects the posterior probability after conditioning on other signals in the locus. Discussion Here we highlight the value of using existing, publicly available data to conduct a multi-stage, multi-trait analysis of related complex traits to understand shared genetic architecture for atherosclerosis and select risk factor traits. Furthermore, this approach allowed us to identify novel pleiotropic susceptibility loci for atherosclerosis. Specifically, we identified 195 shared loci that were novel for atherosclerosis and met our experiment-wide significance threshold, all of which underscore specific underlying pathways linked to subclinical atherosclerosis and an associated risk factor. Multi-trait colocalization further confirmed shared causal signals between atherosclerosis and selected risk factors at 25 novel atherosclerosis loci. Additionally, we integrated gene expression and eQTL data from GTEx to refine the multi-trait signals and identify functional insights for candidate genes involved in the underlying atherosclerosis pathogenesis. Our study underscores the importance of leveraging multi-trait analysis for complex phenotypes like CAD and measures of subclinical atherosclerosis. It is well-established in the literature that CAD shares genomic risk factors with related biological traits and diseases, such as SBP and T2D. 9 , 12 , 13 , 24 While single-trait GWAS continues to identify novel loci, additional resources and follow-up analyses are needed to contextualize newly discovered genetic variants. Our analysis identified 195 novel loci for atherosclerosis, 94 of which were associated with CAD, subclinical atherosclerosis, and their respective risk factors, providing further evidence for their potential roles in clinical disease. Additionally, 101 loci were identified only in analyses with CIMT or CAC and respective risk factors, suggesting a specific role in the early stages of atherosclerosis development. These findings highlight the complementary utility of subclinical traits like CIMT and CAC in uncovering potential novel genetic pathways that precede clinical disease. 19 Importantly, we identified potential regulatory roles involved in the development of atherosclerosis by integrating tissue-specific gene expression data with novel atherosclerosis loci. We identified colocalization between the multi-trait GWAS and tissue-specific gene expression levels at 4 novel loci, involving 5 genes, including SCAI , PRRX2 , and CLIC4 . SCAI expression in arterial tissue colocalized with CIMT, and SBP. SCAI is a negative regulator of Rho protein activation, particularly in the RhoA/DIAPH1 pathway. 34 Prior studies indicate that DIAPH1 knockout in mice attenuates atherosclerosis progression, and downregulation of DIAPH1 expression has been observed in ischemic stroke patients. 35 – 37 These findings suggest that SCAI may influence both structural and functional aspects of vascular biology, warranting further investigation as a potential target for atherosclerosis research. Similarly, CIMT and SBP also colocalized with PRRX2 expression in central and peripheral arterial tissues. PRRX2 is a transcription factor involved in vascular smooth muscle cell differentiation and migration, with established roles in cardiovascular development during embryogenesis. 38 A recent study linked the upregulation of PRRX2 signaling to cardiac remodeling post-myocardial infarction in a mouse model, indicating its potential involvement in SBP and CIMT through arterial vascular smooth muscle cell proliferation or remodeling. We also identified a shared causal signal involving CLIC4 expression in the left ventricle and visceral omentum adipose tissue, that was associated with multi-trait signals including CAD-IMT-SBP and CAD-CAC-SBP. CLIC4 is implicated in apoptosis and inflammation processes that are critical to atherosclerosis development. 39 – 41 Its significant colocalization with CAC, CAD, CIMT, and SBP suggests it may be a central regulator of cardiovascular and metabolic health. Emerging research using in vitro atherosclerosis cell models highlights CLIC4's critical role in endothelial cell function and its potential as a therapeutic target for atherosclerosis. 39 – 42 Future studies should explore the mechanistic pathways of CLIC4 in immune and metabolic regulation and its therapeutic potential for atherosclerosis and hypertension. Our study represents the first multi-trait GWAS to integrate both clinical and subclinical atherosclerosis, leveraging summary statistics from CAD, CAC, CIMT, SBP, LDL-C and T2D to enhance power for detecting novel loci associated with both early and late stages of atherosclerosis progression. This approach provides valuable insights into genetic mechanisms with potential implications for early prevention. By incorporating the largest available GWAS datasets for these traits, we offer a comprehensive perspective on their shared genetic architecture. The multi-trait framework enables the identification of pleiotropic genetic effects, refining risk loci with greater precision and uncovering shared pathways that contribute to atherosclerosis. Additionally, by integrating colocalization and functional genomic analyses, our study provides deeper biological insights linking genetic variants to gene expression and potential causal mechanisms. Addressing heterogeneity in disease progression through the inclusion of both subclinical and clinical phenotypes, our approach captures a broader spectrum of atherosclerosis development, revealing novel insights into genetic factors contributing to both early and late-stage disease, paving the way for potential early intervention and personalized prevention strategies. Nevertheless, this our study is not without limitations. While the GWAS summary statistics include multiple population groups, non-European populations are underrepresented. Cross-population genetic correlations required the use of population-specific reference panels for genetic covariance calculations. Our sensitivity analyses demonstrated similar genetic correlations between cross-population and primarily European GWAS studies, emphasizing European ancestry representation in our dataset. Future studies would benefit from using more diverse LD reference panels and genetic correlation methods that account for multiple genetically inferred genetic ancestral groups. Additionally, the power of our multi-trait colocalization analysis was limited by differences in LD patterns, likely stemming from varying ancestry distributions in the summary statistics. Furthermore, the GWAS for subclinical traits had smaller sample sizes and larger standard errors compared to the GWAS for CAD, LDL-C, T2D, and SBP datasets. Consequently, these disparities in data quality potentially affected the multi-trait colocalization analysis, highlighting the need for larger and more ancestrally balanced datasets for future studies. Finally, we recognize that our analyses was limited to a select number of biological risk factors and may lead to an over-identification of specific biological pathways driven by SBP, T2D, and LDL-C while underrepresenting others. We selected traits based on the strength of their known relationships with not only CAD, but CAC, and CIMT, and prioritized large, well-conducted GWAS. Future studies could benefit from evaluating additional biological risk factors or disease endpoints with subclinical atherosclerosis measures to emphasize additional pathways to atherosclerosis. In summary, our analyses identified novel loci and pathways involved in the development of atherosclerosis, underscoring the importance of subclinical traits like CIMT and CAC in uncovering early-stage mechanisms and the critical role of SBP in CAD development. Multi-trait integration revealed shared causal signals tied to vascular remodeling, inflammation, and metabolic regulation, implicating genes such as SCAI , PRRX2 , and CLIC4 as promising candidates for further research. These findings highlight the value of combining subclinical and clinical traits from publicly available GWAS data to bridge early disease processes with clinical outcomes. Future studies should replicate these discoveries in diverse populations, address ancestry-related gaps, and leverage larger datasets to enhance discovery. Nevertheless, this integrative novel approach holds promise for advancing our understanding of CAD pathogenesis and identifying novel therapeutic targets. Methods We conducted a multi-stage, multi-trait analysis using publicly available GWAS summary statistics for clinical and subclinical atherosclerosis and select cardiometabolic risk factors. We leveraged the largest available CAD, CAC, and CIMT GWAS studies, which included cross-population and European ancestry GWAS, and conducted a multi-stage approach. We conducted multi-trait colocalization analysis across traits and with GTEX v8 (Fig. 1 ). Genetic Association Studies Published GWAS summary statistics from 3 different GWAS on atherosclerosis traits (CAD, CAC, and CIMT) and 3 biological cardiometabolic risk factors (T2D, LDL-C, SBP) were collected. 11 , 12 , 14 , 16 , 19 We selected traits for inclusion in the multi-trait analysis based on their biological relevance to atherosclerosis, the strength of their relationships with atherosclerosis, and the availability of large, high-quality GWAS datasets. To capture complementary aspects of subclinical atherosclerosis across vascular beds, we included CIMT and CAC. Information about each GWAS is available in Supplemental Table 1 , and further details regarding the outcome measure and methods specific to each GWAS are included in the respective publications. All GWAS included well-imputed (r 2 > 0.3) low-frequency or common variants (minor allele frequency > 1%) using either the 1000 Genomes Project, Haplotype Reference Consortium or TOPMed reference panel. 43 – 45 All GWAS consisted of cross-population meta-analyses, with a majority of individuals representing individuals of European ancestry, except for the exception of the SBP GWAS. The SBP GWAS available at the time of this analysis was solely conducted in individuals with European ancestry. 12 Genetic correlation Cross-trait LD score regression using LD Score v1.0 (LDSC) was used to estimate the sample overlap, genetic correlation, and shared heritability for each trait pair. LDSC evaluates genetic correlation and heritability from GWAS summary statistics using a linkage disequilibrium (LD) reference panel. We conducted the analysis using a European LD reference panel made available from the 1000 Genomes Project. 27 , 28 , 46 As a sensitivity analysis, we repeated cross-trait LD score regression for studies that also had summary statistics available that were restricted to European populations. Multi-trait analysis We conducted a multi-stage, multi-trait analysis using N-GWAMA. N-GWAMA applies the estimates from cross-trait LD-score regression to re-weight test statistics from single-trait GWAS studies, by sample size and estimated heritability, while adjusting for genetic covariance across traits. This method enhances the detection of loci with pleiotropic associations by leveraging shared genetic architecture across traits. The newly weighted test statistics are then summated and standardized to produce a new test statistic, from which a new p-value can be calculated for each genetic variant. 27 , 46 We conducted 15 multi-trait analyses using various trait combinations (Fig. 1 ) to identify which combination of traits led to the largest increase in associated variants with atherosclerosis and mitigate any potential power loss from the inclusion of unassociated phenotypes. Stage 1 analyses were restricted to clinical (CAD) and subclinical (CAC and CIMT) atherosclerosis traits. Stage 2 analyses included subclinical atherosclerosis traits with associated risk factors (SBP, LDL-C, T2D). Stage 3 analyses reexamined Stage 2 analyses with the inclusion of CAD (Fig. 1 ). Functional mapping and annotation We used Functional Mapping and Annotation of GWAS (FUMA) to annotate the summary results generated from N-GWAMA. 29 First, variants that met the GWAS significance P-value threshold (P < 5×10 − 8 ) were initially clumped (r 2 < 0.6) using a non-ancestry specific LD reference panel for 1000 Genome Phase 3 to create a genomic risk locus. Within the locus, the variants were clumped again (r 2 < 0.1) to identify independent signals. Lead variants were identified as a subset of the independent significant variants with the strongest signal, while other variants were considered secondary signals. If a lead variant was within 250 kb of another lead variant, the loci were combined. Therefore, a genomic risk locus could contain multiple lead variants. Within FUMA, ANNOVAR was used to map variants to genes by position and annotate variants according to predicted consequence and location. 47 We reported a genomic risk locus if the lead variant from FUMA had P < 0.05 in the included single-trait GWAS summary statistics. A genomic risk locus was considered novel for atherosclerosis and the associated risk factor if the lead variant did not reach the GWAS significance P-value threshold in any of the included single-trait files. A genomic risk locus was considered novel for atherosclerosis if the lead variant did not meet the GWAS genome-wide significance threshold in the single-trait atherosclerosis summary statistics and was more than was more than ± 500 kb away from a GWAS genome-wide significant variant in the GWAS Catalog. When the analysis was restricted to subclinical atherosclerosis traits, a locus was only considered novel for atherosclerosis if it was more than ± 500 kb away from a GWAS genome-wide significant variant for CAD as well. A lead variant was considered to meet experiment-wide significance if it satisfied the Bonferroni-corrected threshold of P < 3.3×10 − 9 (5×10 − 8 /15). To identify shared and unique loci across all multi-trait analyses, we merged loci into a single shared genomic locus if the lead variants from different analyses were within 500 kb of each other. For each shared locus, the variant with the most significant P-value across analyses was designated as the shared lead variant. Shared genomic locus IDs were retained to facilitate the identification of shared regions for colocalization analyses. Trait-trait and Trait- eQTL colocalization Multi-trait colocalization was conducted using HyPrColoc, a Bayesian divisive clustering algorithm designed to identify shared causal signals within a genomic region. 22 HyPrColoc also allows the clustering algorithm to be disabled, enabling colocalization analysis across pre-defined subsets of traits, similar to COLOC and MOLOC. 31 First, we performed trait-trait colocalization without clustering to replicate the multi-stage, multi-trait analysis conducted in N-GWAMA. Single-trait GWAS summary statistics were subset using a ± 500 kb window centered on the lead variant identified by N-GWAMA. Colocalization was then restricted to the traits included in the multi-trait GWAS analysis that identified the lead variant. Second, we conducted trait-trait colocalization across all traits using HyPrColoc’s clustering algorithm, focusing on the lead variant from the shared genomic locus. This analysis applied the variant-specific priors models with default parameters. Evidence for colocalization was determined based on the default variant- -specific regional and alignment priors (P R ∗=P A ∗ =0.5), with colocalization identified when P R P A ≥0.25. Strong evidence of colocalization across traits was defined as P R P A ≥ 0.80. Finally, for loci that colocalized in the trait-trait analysis, we performed trait-eQTL colocalization using eQTL data from GTEx v8 using the COLOC package in R. 31 ,32 33, We leveraged summary statistics from the multi-trait GWAS for corresponding trait-trait colocalization and focused our analysis on a subset of GTEx v8 tissues selected based on their biological relevance to the studied traits ( Supplemental Table 8) . Colocalization was assessed using the default prior probabilities. Evidence for colocalization was defined as a posterior probability of a shared causal variant (PP H4 ) ≥ 0.50 and a conditional posterior probability (PP c =PP H4 /(PP H3 +PP H4) ) ≥ 0.80. Strong evidence for colocalization was defined as (PP H4 ) ≥ 0.80 and a conditional posterior probability (PP H4 /(PP H3 +PP H4 )) ≥ 0.80. Declarations Conflict of interest Dr. Clint L. Miller has received funding from AstraZeneca for unrelated work. All other authors have no conflicts of interest to declare. Acknowledgements This study was funded by the National Institutes of Health (NIH) National Heart, Lung, and Blood Institute (NHLBI) grant number R01 HL146860 to P.S.dV. We thank GLGLC, ICBP and DIAMANTE consortia for making their summary statistics available to the research community. We also acknowledge funding support from NIH grants (R01HL148239, R01HL164577 and U01DK142283), Leducq Foundation network grant ‘COMET’ (24CVD02), and American Heart Association Transformational Project Award (24TPA1300556) to C.L.M. References Klarin D et al (2017) Genetic analysis in UK Biobank links insulin resistance and transendothelial migration pathways to coronary artery disease. Nat Genet 49:1392 Zhuang Z, Yao M, Wong JYY, Liu Z, Huang T (2021) Shared genetic etiology and causality between body fat percentage and cardiovascular diseases: a large-scale genome-wide cross-trait analysis. BMC Med 19:100 Tada H et al (2014) Multiple associated variants increase the heritability explained for plasma lipids and coronary artery disease. Circ Cardiovasc Genet 7:583–587 Deloukas P et al (2013) Large-scale association analysis identifies new risk loci for coronary artery disease. Nat Genet 45:25–33 Howson JMM et al (2017) Fifteen new risk loci for coronary artery disease highlight arterial-wall-specific mechanisms. Nat Genet 49:1113–1119 O'Donnell CJ et al (2011) Genome-wide association study for coronary artery calcification with follow-up in myocardial infarction. Circulation 124:2855–2864 Webb TR et al (2017) Systematic Evaluation of Pleiotropy Identifies 6 Further Loci Associated With Coronary Artery Disease. J Am Coll Cardiol 69:823–836 Vinkhuyzen AAE, Wray NR, Yang J, Goddard ME, Visscher PM (2013) Estimation and Partition of Heritability in Human Populations Using Whole-Genome Analysis Methods. Annu Rev Genet 47:75–95 LeBlanc M et al (2016) Identifying Novel Gene Variants in Coronary Artery Disease and Shared Genes With Several Cardiovascular Risk Factors. Circ Res 118:83–94 Aragam KG et al (2022) Discovery and systematic characterization of risk variants and genes for coronary artery disease in over a million participants. Nat Genet 54:1803–1815 Tcheandjieu C et al (2022) Large-scale genome-wide association study of coronary artery disease in genetically diverse populations. Nat Med 28:1679–1692 Evangelou E et al (2018) Genetic analysis of over 1 million people identifies 535 new loci associated with blood pressure traits. Nat Genet 50:1412–1425 Warren HR et al (2017) Genome-wide association analysis identifies novel blood pressure loci and offers biological insights into cardiovascular risk. Nat Genet 49:403–415 Graham SE et al (2021) The power of genetic diversity in genome-wide association studies of lipids. Nature 600:675–679 Yeung MW et al (2022) Twenty-Five Novel Loci for Carotid Intima-Media Thickness: A Genome-Wide Association Study in > 45 000 Individuals and Meta-Analysis of > 100 000 Individuals. Arterioscler Thromb Vasc Biol 42:484–501 Mahajan A et al (2022) Multi-ancestry genetic study of type 2 diabetes highlights the power of diverse populations for discovery and translation. Nat Genet 54:560–572 Morris AP et al (2012) Large-scale association analysis provides insights into the genetic architecture and pathophysiology of type 2 diabetes. Nat Genet 44:981–990 Scott RA et al (2017) An Expanded Genome-Wide Association Study of Type 2 Diabetes in Europeans. Diabetes 66:2888–2902 Kavousi M et al (2023) Multi-ancestry genome-wide study identifies effector genes and druggable pathways for coronary artery calcification. Nat Genet 55:1651–1664 Meena D et al (2025) Genome-wide association study and multi-ancestry meta-analysis identify common variants associated with carotid artery intima-media thickness. medRxiv , 04.11.25325582 (2025) Dai X, Wiernek S, Evans JP, Runge MS (2016) Genetics of coronary artery disease and myocardial infarction. World J Cardiol 8:1–23 Foley CN et al (2021) A fast and efficient colocalization algorithm for identifying shared genetic risk factors across multiple traits. Nat Commun 12:764 Baselmans BML et al (2019) Multivariate genome-wide analyses of the well-being spectrum. Nat Genet 51:445–451 Zhao W et al (2017) Identification of new susceptibility loci for type 2 diabetes and shared etiological pathways with coronary heart disease. Nat Genet 49:1450–1457 Siewert KM, Voight BF (2018) Bivariate Genome-Wide Association Scan Identifies 6 Novel Loci Associated With Lipid Levels and Coronary Artery Disease. Circ Genom Precis Med 11:e002239 Levin MG et al (2022) Genome-wide association and multi-trait analyses characterize the common genetic architecture of heart failure. Nat Commun 13:6914 Bulik-Sullivan BK et al (2015) LD Score regression distinguishes confounding from polygenicity in genome-wide association studies. Nat Genet 47:291–295 Auton A et al (2015) A global reference for human genetic variation. Nature 526:68–74 Watanabe K, Taskesen E, van Bochoven A, Posthuma D (2017) Functional mapping and annotation of genetic associations with FUMA. Nat Commun 8:1826 Cerezo M et al (2024) The NHGRI-EBI GWAS Catalog: standards for reusability, sustainability and diversity. Nucleic Acids Res 53:D998–D1005 Giambartolomei C et al (2018) A Bayesian framework for multiple trait colocalization from summary association statistics. Bioinformatics 34:2538–2545 Barbeira AN et al (2020) Exploiting the GTEx resources to decipher the mechanisms at GWAS loci. bioRxiv, 814350 The Genotype-Tissue (2013) Expression (GTEx) project. Nat Genet 45:580–585 Fabregat A et al (2017) Reactome pathway analysis: a high-performance in-memory approach. BMC Bioinformatics 18:142 Senatus L et al (2023) DIAPH1 mediates progression of atherosclerosis and regulates hepatic lipid metabolism in mice. Commun Biol 6:280 Ren Z et al (2020) Association of DIAPH1 gene polymorphisms with ischemic stroke. Aging 12:416–435 Ramasamy R, Shekhtman A, Schmidt AM (2023) RAGE/DIAPH1 and atherosclerosis through an evolving lens: Viewing the cell from the Inside - Out. Atherosclerosis, 117304 Jones FS et al (2002) Regulation of vascular smooth muscle cell growth and adhesion by paired-related homeobox genes. Chest 121:89s–90s Zhang X et al (2020) MicroRNA-217-5p ameliorates endothelial cell apoptosis induced by ox-LDL by targeting CLIC4. Nutr Metabolism Cardiovasc Dis 30:523–533 Higashi Y (2022) Roles of Oxidative Stress and Inflammation in Vascular Endothelial Dysfunction-Related Disease. in Antioxidants Vol. 11 Kleinjan ML et al (2023) CLIC4 Regulates Endothelial Barrier Control by Mediating PAR1 Signaling via RhoA. Arterioscler Thromb Vasc Biol 43:1441–1454 Shao X, Liu Z, Liu S, Lin N, Deng Y (2021) Astragaloside IV alleviates atherosclerosis through targeting circ_0000231/miR-135a-5p/CLIC4 axis in AS cell model in vitro. Mol Cell Biochem 476:1783–1795 Taliun D et al (2021) Sequencing of 53,831 diverse genomes from the NHLBI TOPMed Program. Nature 590:290–299 Chou W-C et al (2016) A combined reference panel from the 1000 Genomes and UK10K projects improved rare variant imputation in European and Chinese samples. Sci Rep 6:39313 McCarthy S et al (2016) A reference panel of 64,976 haplotypes for genotype imputation. Nat Genet 48:1279–1283 Bulik-Sullivan B et al (2015) An atlas of genetic correlations across human diseases and traits. Nat Genet 47:1236–1241 Wang K, Li M, Hakonarson H (2010) ANNOVAR: functional annotation of genetic variants from high-throughput sequencing data. Nucleic Acids Res 38:e164–e164 Additional Declarations Yes there is potential Competing Interest. Dr. Clint L. Miller has received funding from AstraZeneca for unrelated work. All other authors have no conflicts of interest to declare. Supplementary Files SupplementalMaterialsNatureCardioHasbanietal2025.pdf Supplemental Table for Mulit-trait analyses of clinical and subclinical atherosclerosis SupplementalFiguresNatureCardioHasbanietal2025.docx Supplemental Figures for Multi-trait analysis of clinical and subclinical atherosclerosis Cite Share Download PDF Status: Posted Version 1 posted You are reading this latest preprint version Research Square lets you share your work early, gain feedback from the community, and start making changes to your manuscript prior to peer review in a journal. As a division of Research Square Company, we’re committed to making research communication faster, fairer, and more useful. We do this by developing innovative software and high quality services for the global research community. Our growing team is made up of researchers and industry professionals working together to solve the most critical problems facing scientific publishing. Also discoverable on Platform About Our Team In Review Editorial Policies Advisory Board Help Center Resources Author Services Accessibility API Access RSS feed Manage Cookie Preferences © Research Square 2026 | ISSN 2693-5015 (online) Privacy Policy Terms of Service Do Not Sell My Personal Information {"props":{"pageProps":{"initialData":{"identity":"rs-6456056","acceptedTermsAndConditions":true,"allowDirectSubmit":true,"archivedVersions":[],"articleType":"Article","associatedPublications":[],"authors":[{"id":444840976,"identity":"4a34b68f-3304-4767-8137-50ce1167177f","order_by":0,"name":"Natalie R. Hasbani","email":"","orcid":"https://orcid.org/0000-0001-9874-3440","institution":"Human Genetics Center, Department of Epidemiology, School of Public Health, The University of Texas Health Science Center at Houston, Houston, Texas, USA","correspondingAuthor":false,"prefix":"","firstName":"Natalie","middleName":"R.","lastName":"Hasbani","suffix":""},{"id":444840977,"identity":"e2eeb926-7d3f-4b89-8a03-c2cc05e3f35f","order_by":1,"name":"Adam S. Heath","email":"","orcid":"https://orcid.org/0000-0002-5909-9055","institution":"Human Genetics Center, Department of Epidemiology, School of Public Health, The University of Texas Health Science Center at Houston, Houston, Texas, USA","correspondingAuthor":false,"prefix":"","firstName":"Adam","middleName":"S.","lastName":"Heath","suffix":""},{"id":444840978,"identity":"c96de596-89b5-4995-9cba-c551e04b0963","order_by":2,"name":"Chani J. Hodonsky","email":"","orcid":"","institution":"Department of Genome Sciences, University of Virginia, Charlottesville, Virginia, USA","correspondingAuthor":false,"prefix":"","firstName":"Chani","middleName":"J.","lastName":"Hodonsky","suffix":""},{"id":444840979,"identity":"80bc7cf9-57c2-4d58-bb45-fff9cb2b4da0","order_by":3,"name":"Julie Hahn","email":"","orcid":"","institution":"Human Genetics Center, Department of Epidemiology, School of Public Health, The University of Texas Health Science Center at Houston, Houston, Texas, USA","correspondingAuthor":false,"prefix":"","firstName":"Julie","middleName":"","lastName":"Hahn","suffix":""},{"id":444840980,"identity":"8e886c40-6629-4a0b-9f78-7f6b9a560493","order_by":4,"name":"Devendra Meena","email":"","orcid":"","institution":"Department of Biostatistics and Epidemiology, MRC-PHE Centre for Environment and Health, School of Public Health, Imperial College London, London UK","correspondingAuthor":false,"prefix":"","firstName":"Devendra","middleName":"","lastName":"Meena","suffix":""},{"id":444840981,"identity":"64893a41-04ed-4255-990a-15da672f6864","order_by":5,"name":"Haojie Lu","email":"","orcid":"","institution":"Department of Epidemiology, Erasmus University Medical Center, Rotterdam, the Netherlands","correspondingAuthor":false,"prefix":"","firstName":"Haojie","middleName":"","lastName":"Lu","suffix":""},{"id":444840982,"identity":"92ac85f9-5c42-4602-8ddf-8f5e5d3632f3","order_by":6,"name":"Abbas Dehghan","email":"","orcid":"","institution":"Department of Biostatistics and Epidemiology, MRC-PHE Centre for Environment and Health, School of Public Health, Imperial College London, London UK; MRC Centre for Environment and Health, Imperial College London, London, UK;\tUK Dementia Research Institute Centre at Imperial College London, London, UK","correspondingAuthor":false,"prefix":"","firstName":"Abbas","middleName":"","lastName":"Dehghan","suffix":""},{"id":444840983,"identity":"d8bac5ab-8dd9-4e63-8671-2e7fb06957e4","order_by":7,"name":"Maryam Kavousi","email":"","orcid":"https://orcid.org/0000-0001-5976-6519","institution":"Department of Epidemiology, Erasmus University Medical Center, Rotterdam, the Netherlands","correspondingAuthor":false,"prefix":"","firstName":"Maryam","middleName":"","lastName":"Kavousi","suffix":""},{"id":444840984,"identity":"2c1987fc-69bc-437d-86f0-cf71ef4fc3ef","order_by":8,"name":"Benjamin F. Voight","email":"","orcid":"https://orcid.org/0000-0002-6205-9994","institution":"Institute for Biomedical Informatics, University of Pennsylvania, Philadelphia, USA; Department of Genetics, Perelman School of Medicine, University of Pennsylvania, Philadelphia Pennsylvania, USA; Department of Systems Pharmacology and Translational Therapeutics, University of Pennsylvania, Philadelphia Pennsylvania, USA; Institute of Translational Medicine and Therapeutics, Perelman School of Medicine, University of Pennsylvania, Philadelphia, Pennsylvania, USA","correspondingAuthor":false,"prefix":"","firstName":"Benjamin","middleName":"F.","lastName":"Voight","suffix":""},{"id":444840985,"identity":"73a91812-b91d-4559-a7d0-48220f08dc28","order_by":9,"name":"Patricia A. Peyser","email":"","orcid":"https://orcid.org/0000-0002-9717-8459","institution":"Department of Epidemiology, School of Public Health, University of Michigan, Ann Arbor, MI, USA","correspondingAuthor":false,"prefix":"","firstName":"Patricia","middleName":"A.","lastName":"Peyser","suffix":""},{"id":444840986,"identity":"5e27e785-d6b6-48d6-8ca7-e065a3edf532","order_by":10,"name":"Alanna C. Morrison","email":"","orcid":"https://orcid.org/0000-0001-6381-4296","institution":"Human Genetics Center, Department of Epidemiology, School of Public Health, The University of Texas Health Science Center at Houston, Houston, Texas, USA","correspondingAuthor":false,"prefix":"","firstName":"Alanna","middleName":"C.","lastName":"Morrison","suffix":""},{"id":444840987,"identity":"87e2eb83-d308-40e5-b87a-1c77c147f1d4","order_by":11,"name":"Themistocles L. Assimes","email":"","orcid":"https://orcid.org/0000-0003-2349-0009","institution":"VA Palo Alto Health Care System, Palo Alto, CA, USA; Department of Medicine, Stanford University School of Medicine, Stanford, CA, USA; Cardiovascular Institute, Stanford University School of Medicine, Stanford, CA, United States; Department of Epidemiology and Population Health, Stanford University School of Medicine, Stanford, CA, United States","correspondingAuthor":false,"prefix":"","firstName":"Themistocles","middleName":"L.","lastName":"Assimes","suffix":""},{"id":444840988,"identity":"73822332-a2d9-422a-ac4c-8e6ba1ee1b48","order_by":12,"name":"Scott M. Damrauer","email":"","orcid":"https://orcid.org/0000-0001-8009-1632","institution":"Institute for Biomedical Informatics, University of Pennsylvania, Philadelphia, USA; Department of Genetics, Perelman School of Medicine, University of Pennsylvania, Philadelphia Pennsylvania, USA; Department of Surgery, Perelman School of Medicine, University of Pennsylvania, Philadelphia, Pennsylvania, USA;\tCardiovascular Institute, University of Pennsylvania Perelman School of Medicine, Philadelphia; Corporal Michael Crescenz VA Medical Center, Philadelphia, Pennsylvania, USA","correspondingAuthor":false,"prefix":"","firstName":"Scott","middleName":"M.","lastName":"Damrauer","suffix":""},{"id":444840989,"identity":"bae19538-266a-47a7-a1d7-0edb66fe6f5c","order_by":13,"name":"Clint L. Miller","email":"","orcid":"https://orcid.org/0000-0003-4276-3607","institution":"Department of Genome Sciences, University of Virginia, Charlottesville, Virginia, USA; Department of Biochemistry and Molecular Genetics, University of Virginia, Charlottesville, VA, USA; Department of Public Health Sciences, University of Virginia, Charlottesville, VA, USA","correspondingAuthor":false,"prefix":"","firstName":"Clint","middleName":"L.","lastName":"Miller","suffix":""},{"id":444840975,"identity":"f1f5b351-81d3-4618-be74-41f8bf996ffa","order_by":14,"name":"Paul S. de Vries","email":"data:image/png;base64,iVBORw0KGgoAAAANSUhEUgAAAZAAAAAyAQMAAABI0h/eAAAABlBMVEX///8AAABVwtN+AAAACXBIWXMAAA7EAAAOxAGVKw4bAAABC0lEQVRIie3Rv0rEMBzA8V8Qest5rhGtvkKOrkVfJaHgVMEHODBSiEvl1iu+hD7B/Y5Cu1RdCy6ng5NDweWGDKZ3XF0a7kaHfLf8+ZCEALhc/7BLCQRBA5GUAwKEAHS7hv2EmXkkqiNXMKRmsJvIDTHle5CyYi3xp8OXD2z0m5ifyMFnPYGzUc3735LGrH1LkKUlW2TqXaSnSJK4gODYQhi0xAPxVBWQH0pDKCfJtTQzNnL0vSHzUkGu9WtHbq2Ebk8ZGAIedoQzK/m6QaFoMMs9WDyoKEhrcfcYF3ScVUvLxaLnZaNDf5p4B81KX/j3swh/4kl4Pir7T1nH/368bf0ttH+ry+VyufbqF0qra8YigvAiAAAAAElFTkSuQmCC","orcid":"https://orcid.org/0000-0003-0964-0111","institution":"Human Genetics Center, Department of Epidemiology, School of Public Health, The University of Texas Health Science Center at Houston, Houston, Texas, USA","correspondingAuthor":true,"prefix":"","firstName":"Paul","middleName":"S.","lastName":"de Vries","suffix":""}],"badges":[],"createdAt":"2025-04-15 14:56:21","currentVersionCode":1,"declarations":"","doi":"10.21203/rs.3.rs-6456056/v1","doiUrl":"https://doi.org/10.21203/rs.3.rs-6456056/v1","draftVersion":[],"editorialEvents":[],"editorialNote":"","failedWorkflow":false,"files":[{"id":80995342,"identity":"f74f61af-00c7-495f-821f-90f46770baf1","added_by":"auto","created_at":"2025-04-21 05:03:36","extension":"jpeg","order_by":1,"title":"Figure 1","display":"","copyAsset":false,"role":"figure","size":134191,"visible":true,"origin":"","legend":"\u003cp\u003e\u003cstrong\u003eOverview of multi-stage, multi-trait genome-wide association study\u003c/strong\u003e\u003c/p\u003e\n\u003cp\u003eFigure 1 provides an overview of the multi-stage, multi-trait genome-wide association study. Stage 1 trait combinations were coronary artery disease (CAD)-coronary artery calcification (CAC), CAD-carotid intima- media thickness (CIMT), and CAC-CIMT. Stage 2 trait combinations were CAC or CIMT with systolic blood pressure (SBP), low- density lipoprotein cholesterol (LDL-C), or type 2 diabetes (T2D). Stage 3 trait combinations were CAD and CAC or CAD and CIMT with SBP, LDL-C or T2D. We first conducted linkage disequilibrium score regression using LD Score (LDSC) v1, then conducted 15 multi-trait genome-wide association analyses in N-weighted multivariate genome-wide association meta-analysis (N-GWAMA), annotated using Functional Mapping and Annotation of GWAS (FUMA) to define risk loci, prior to performing trait-trait colocalization in HyPrColoc and trait-eQTL colocalization using COLOC packages in R.\u003c/p\u003e\n\u003cp\u003eAbbreviations: CAC, coronary artery calcification; CAD, coronary artery disease; CIMT,carotid intima- media thickness; FUMA, Functional Mapping and Annotation of GWAS; eQTL, expression quantitative loci; \u0026nbsp;GWAS, genome-wide association study; LDL-C, low- density lipoprotein cholesterol; N-weighted multivariate genome-wide association meta-analysis,N-GWAMA; SBP, systolic blood pressure, SNP, single nucleotide polymorphism\u003c/p\u003e","description":"","filename":"image1.jpeg","url":"https://assets-eu.researchsquare.com/files/rs-6456056/v1/013afb6f900755c63dc63c25.jpeg"},{"id":80995344,"identity":"af4b2133-7733-48d9-b48f-049461419d97","added_by":"auto","created_at":"2025-04-21 05:03:36","extension":"png","order_by":2,"title":"Figure 2","display":"","copyAsset":false,"role":"figure","size":131562,"visible":true,"origin":"","legend":"\u003cp\u003e\u003cstrong\u003eGenetic correlations across atherosclerosis and associated risk factors\u003c/strong\u003e\u003c/p\u003e\n\u003cp\u003eFigure 2 is a heatmap of genetic correlation coefficients across atherosclerosis and risk factor traits. Dark red refers to a strong positive genetic correlation while dark blue indicates no correlation. Asterisk indicates a significant genetic correlation (P\u0026lt;0.05). Panel A provides genetic correlation coefficients using cross-population summary statistics for coronary artery disease (CAD), carotid itima media thickness (CIMT), coronary artery calcification (CAC) , low density lipoprotein cholesterol (LDL-C) and type 2 diabetes (T2D). Panel B provides European-specific genetic correlation coefficients for studies as were provided for CAC, CIMT, LDL-C, systolic blood pressure.\u003c/p\u003e\n\u003cp\u003eAbbreviations: CAC, coronary artery calcification; CAD, coronary artery disease; CIMT, carotid intima media thickness, LDL-C, low density lipoprotein cholesterol; SBP, systolic blood pressure, T2D, type 2 diabetes\u003c/p\u003e","description":"","filename":"image2.png","url":"https://assets-eu.researchsquare.com/files/rs-6456056/v1/2f76ee0c638926204e585736.png"},{"id":80997501,"identity":"598f4b25-8c8a-431d-bf64-ef45546b8934","added_by":"auto","created_at":"2025-04-21 05:35:36","extension":"png","order_by":3,"title":"Figure 3","display":"","copyAsset":false,"role":"figure","size":247891,"visible":true,"origin":"","legend":"\u003cp\u003e\u003cstrong\u003eSummary of shared risk loci identified in multi-trait genome-wide association analysis\u003c/strong\u003e\u003c/p\u003e\n\u003cp\u003eFigure 3 is an upset plot summarizing the shared loci identified in our multi-trait genome- wide association study (GWAS). The y-axis contains the counts for the number of shared combined genetic risk loci across stages of analysis. Each individual\u0026nbsp; risk locus was identified in a multi-trait analysis of atherosclerosis and select risk factor traits. Lead variants from an individual genetic risk loci were collapsed into a single shared risk locus if lead variants were within 500 kb of each other. Lines connect groups of traits within a shared risk locus. A locus was considered novel for atherosclerosis if the lead variant did not meet the GWAS significance threshold (P=5×10\u003csup\u003e-8\u003c/sup\u003e) in the single-trait atherosclerosis summary statistics and was more than was ±500 kb away from a GWAS significant variant in the GWAS Catalog. When the analysis was restricted to subclinical atherosclerosis traits, a locus was considered novel for atherosclerosis if it was more than ±500 kb away from a GWAS significant variant for coronary artery disease.\u003c/p\u003e\n\u003cp\u003eAbbreviations: CAC, coronary artery calcification; CAD, coronary artery disease; CIMT, carotid intima- media thickness; GWAS, genome-wide association study; LDL-C, low- density lipoprotein cholesterol; SBP, systolic blood pressure, T2D, type 2 diabetes\u003c/p\u003e","description":"","filename":"image3.png","url":"https://assets-eu.researchsquare.com/files/rs-6456056/v1/c66e958213d3e1d03ad93ff5.png"},{"id":80995349,"identity":"d6aadfec-d796-4f81-966a-28c545a89f0b","added_by":"auto","created_at":"2025-04-21 05:03:36","extension":"png","order_by":4,"title":"Figure 4","display":"","copyAsset":false,"role":"figure","size":172222,"visible":true,"origin":"","legend":"\u003cp\u003e\u003cstrong\u003eNovel atherosclerosis loci with evidence of trait-trait colocalization\u003c/strong\u003e\u003c/p\u003e\n\u003cp\u003eFigure 4 summarizes the multi-trait genome-wide association study (GWAS) and multi-trait colocalization performed to identify novel loci associated with atherosclerosis.\u0026nbsp;Panel A\u0026nbsp;provides the results of multi-trait GWAS. The\u0026nbsp;\u003cem\u003ex\u003c/em\u003e-axis denotes the single trait GWAS and N-weighted genome- -wide association meta-analysis (N-GWAMA) results, and the\u0026nbsp;\u003cem\u003ey\u003c/em\u003e-axis denotes the pleiotropic independent lead variants at each locus with the nearest gene in parenthesis. The presence of a dot indicates a trait involved in the N-GWAMA. The size of each point denotes the absolute\u0026nbsp;\u003cem\u003ez\u003c/em\u003e-score for each trait. Associations exceeding the Bonferroni threshold are denoted with a white circle. Variants are grouped by chromosome. Panel\u0026nbsp;B\u0026nbsp;presents the results of multi-trait colocalization. The\u0026nbsp;\u003cem\u003ex\u003c/em\u003e-axis denotes all atherosclerosis and risk factor traits. The\u0026nbsp;\u003cem\u003ey\u003c/em\u003e-axis represents the lead variant at each independent locus identified in the multi-trait GWAS. Lines connect groups of traits with evidence of colocalization at a given locus. The size of each point represents the posterior probability for colocalization. Evidence for colocalization was determined based on the default variant-specific regional and alignment priors\u0026nbsp;(P\u003csub\u003e\u003cem\u003eR\u003c/em\u003e\u003c/sub\u003e∗=P\u003csub\u003e\u003cem\u003eA\u003c/em\u003e\u003c/sub\u003e\u003cem\u003e∗\u003c/em\u003e=0.5), with colocalization identified when\u0026nbsp;P\u003csub\u003e\u003cem\u003eR\u003c/em\u003e\u003c/sub\u003eP\u003csub\u003e\u003cem\u003eA\u003c/em\u003e\u003c/sub\u003e≥0.25\u003cstrong\u003e. \u003c/strong\u003eResults are restricted to those with P\u003csub\u003e\u003cem\u003eR\u003c/em\u003e\u003c/sub\u003eP\u003csub\u003e\u003cem\u003eA\u003c/em\u003e\u003c/sub\u003eA\u0026gt;0.80.\u003c/p\u003e\n\u003cp\u003eAbbreviations: CAC, coronary artery calcification; CAD, coronary artery disease; CIMT, carotid intima- media thickness; GWAS, genome-wide association study; LDL-C, low- density lipoprotein cholesterol; N-GWAMA, N-weighted genome- -wide association meta-analysis; SBP, systolic blood pressure, T2D, type 2 diabetes\u003c/p\u003e","description":"","filename":"image4.png","url":"https://assets-eu.researchsquare.com/files/rs-6456056/v1/488e4961dbfe34f3614f7fb1.png"},{"id":80995354,"identity":"95edf659-de92-4b5b-bafe-3ddf30f941a1","added_by":"auto","created_at":"2025-04-21 05:03:36","extension":"png","order_by":5,"title":"Figure 5","display":"","copyAsset":false,"role":"figure","size":462583,"visible":true,"origin":"","legend":"\u003cp\u003e\u003cstrong\u003eColocalization for atherosclerosis and low density lipoprotein cholesterol with \u003c/strong\u003e\u003cem\u003e\u003cstrong\u003eBNC2 \u003c/strong\u003e\u003c/em\u003e\u003cstrong\u003eexpression in whole blood\u003c/strong\u003e\u003c/p\u003e\n\u003cp\u003eFigure 5 illustrates the colocalization of expression for\u003cem\u003e BNC2 \u003c/em\u003ein whole blood with all measures of atherosclerosis and low-density lipoprotein cholesterol (LDL-C) highlighting potential shared causal variants at rs respectively. Panels A and B depict scatter plots of the gene expression in whole blood (-log10(p)) against coronary artery disease (CAD), coronary artery calcification (CAC) and LDL-C_ multi-trait GWAS results (-log10(p)), with linkage disequilibrium structure indicated by color coding. Panels C provide locus-specific association plots across CAD, CAC, CIMT GWAS and whole blood expression quantitative trait loci analyses, further supporting colocalization at these loci.\u003c/p\u003e","description":"","filename":"image5.png","url":"https://assets-eu.researchsquare.com/files/rs-6456056/v1/3b792dd5af29b6717b1678a1.png"},{"id":80995357,"identity":"35051662-e909-4a10-8725-eadfada63dde","added_by":"auto","created_at":"2025-04-21 05:03:37","extension":"png","order_by":6,"title":"Figure 6","display":"","copyAsset":false,"role":"figure","size":249805,"visible":true,"origin":"","legend":"\u003cp\u003e\u003cstrong\u003eColocalization for systolic blood pressure and carotid intima media thickness with \u003c/strong\u003e\u003cem\u003e\u003cstrong\u003eSCAI, PRRX2 and PPP6C \u003c/strong\u003e\u003c/em\u003e\u003cstrong\u003eexpression in the Tibial Artery\u003c/strong\u003e\u003c/p\u003e\n\u003cp\u003eFigure 6 illustrates the colocalization of expression for\u003cem\u003e PRRX2, PPP6C \u003c/em\u003eand \u003cem\u003eSCAI \u003c/em\u003e\u0026nbsp;in the tibial artery with systolic blood pressure (SBP) and carotid intima- media thickness (CIMT), highlighting potential shared causal variants at rs11780582 and rs11790512, respectively. Panels A B and \u0026nbsp;C depict scatter plots of the gene expression in Tibial Artery (-log10(p)) against SBP-CIMT multi-trait GWAS (-log10(p)) results, with linkage disequilibrium structure indicated by color coding.\u0026nbsp;\u003c/p\u003e\n\u003cp\u003eAbbreviations: CIMT, C, carotid intima- media thickness; eQTL,expression quantitative trait loci, GWAS, genome-wide association study; SBP, systolic blood pressure\u003c/p\u003e","description":"","filename":"image6.png","url":"https://assets-eu.researchsquare.com/files/rs-6456056/v1/1e9338efee9e585de0136ae7.png"},{"id":80995362,"identity":"99658f5c-53fd-4e90-bd1e-d0528ae503c7","added_by":"auto","created_at":"2025-04-21 05:03:37","extension":"png","order_by":7,"title":"Figure 7","display":"","copyAsset":false,"role":"figure","size":426615,"visible":true,"origin":"","legend":"\u003cp\u003e\u003cstrong\u003eColocalization for SBP, CAC, and CAD with \u003c/strong\u003e\u003cem\u003e\u003cstrong\u003eCLIC4 \u003c/strong\u003e\u003c/em\u003e\u003cstrong\u003eexpression in left ventricle and visceral adipose tissue\u003c/strong\u003e\u003c/p\u003e\n\u003cp\u003eFigure 7 illustrates the colocalization of expression of \u003cem\u003eCLIC4\u003c/em\u003e in left ventricle tissue and visceral adipose tissue with coronary artery disease (CAD), coronary artery calcification (CAC), and systolic blood pressure (SBP), highlighting rs as a potential shared causal variant. Panel A presents a scatter plot comparing gene expression (-log10(p)) in left ventricle heart tissue against SBP-CAC-CAD multi-trait GWAS results, with a, color-coded linkage disequilibrium structure. Panel B provides locus-specific association plots for SBP, CAC, and CAD GWAS, as well as \u003cem\u003eCLIC4\u003c/em\u003e expression quantitative trait loci analyses, showing rs4366267 as a key locus across all datasets.\u003c/p\u003e\n\u003cp\u003eAbbreviations: CAC, coronary artery calcification; CAD, coronary artery disease; eQTL,expression quantitative trait loci; GWAS, genome-wide association study; SBP, systolic blood pressure\u003c/p\u003e","description":"","filename":"image7.png","url":"https://assets-eu.researchsquare.com/files/rs-6456056/v1/3c13c85846e134402854a7de.png"},{"id":84379060,"identity":"be0d45d1-36cb-4849-8e9d-3ba751ec381e","added_by":"auto","created_at":"2025-06-11 08:53:58","extension":"pdf","order_by":0,"title":"","display":"","copyAsset":false,"role":"manuscript-pdf","size":3049873,"visible":true,"origin":"","legend":"","description":"","filename":"manuscript.pdf","url":"https://assets-eu.researchsquare.com/files/rs-6456056/v1/4253a1e6-47dd-46aa-b8e8-7fb8da14000d.pdf"},{"id":80995750,"identity":"6c16048d-e106-406d-83a1-b660200535ea","added_by":"auto","created_at":"2025-04-21 05:11:36","extension":"pdf","order_by":1,"title":"","display":"","copyAsset":false,"role":"supplement","size":1039245,"visible":true,"origin":"","legend":"Supplemental Table for Mulit-trait analyses of clinical and subclinical atherosclerosis","description":"","filename":"SupplementalMaterialsNatureCardioHasbanietal2025.pdf","url":"https://assets-eu.researchsquare.com/files/rs-6456056/v1/b0d31531c10b7bef3385bc55.pdf"},{"id":80995360,"identity":"5e97000b-4677-4d50-b6f3-fbfde55b4dbd","added_by":"auto","created_at":"2025-04-21 05:03:37","extension":"docx","order_by":2,"title":"","display":"","copyAsset":false,"role":"supplement","size":3401045,"visible":true,"origin":"","legend":"Supplemental Figures for Multi-trait analysis of clinical and subclinical atherosclerosis","description":"","filename":"SupplementalFiguresNatureCardioHasbanietal2025.docx","url":"https://assets-eu.researchsquare.com/files/rs-6456056/v1/84d2db6328424d5324158729.docx"}],"financialInterests":"\u003cb\u003eYes\u003c/b\u003e there is potential Competing Interest.\nDr. Clint L. Miller has received funding from AstraZeneca for unrelated work. All other authors have no conflicts of interest to declare.","formattedTitle":"A multi-trait genome-wide association study of coronary artery disease and subclinical atherosclerosis traits","fulltext":[{"header":"Introduction","content":"\u003cp\u003eCoronary artery disease (CAD) is a complex disease process involving a convergence of environmental and genetic risk factors. Several underlying heritable traits exist which indicate the presence of atherosclerosis prior to the clinical manifestation of CAD, including coronary artery calcification (CAC) and carotid intima- media thickness (CIMT). \u0026nbsp;Recent genetic studies also provide abundant evidence of pleiotropy among risk factors with approximately 50% of identified CAD risk loci associating with underlying clinical risk factors.\u003csup\u003e1-11\u003c/sup\u003e\u003c/p\u003e\n\u003cp\u003eGenome-wide association studies (GWAS) for direct and indirect measures of atherosclerosis as well as associated risk factors such as systolic blood pressure (SBP), low-density lipoprotein cholesterol (LDL-C), and type 2 diabetes (T2D) have identified hundreds of susceptibility loci.\u003csup\u003e10-20\u003c/sup\u003e However, the discovered loci collectively only explain a fraction of the heritability of these phenotypes, suggesting that additional associated susceptibility loci remain to be discovered.\u003csup\u003e21\u003c/sup\u003e Combining information from the shared genetic architecture between atherosclerosis measures and clinical risk factors may help identify new susceptibility loci with shared underlying biology. Statistical approaches that facilitate the joint analysis of multiple correlated traits have been developed that increase the power to detect loci that are pleiotropically associated with more than one trait.\u003csup\u003e22,23\u003c/sup\u003e These methods have thus far uncovered many shared susceptibility loci for cardiometabolic phenotypes and provided unique insight into shared genetic pathways.\u003csup\u003e24-26\u003c/sup\u003e\u003c/p\u003e\n\u003cp\u003eNo studies to date have used multi-trait approaches to evaluate both subclinical and clinical atherosclerosis phenotypes along with risk factors. Joint analysis of clinical CAD with subclinical atherosclerosis traits and related risk factors may uncover key susceptibility loci and pleiotropic mechanisms contributing to the development of CAD. Here, we report a multi-stage, multi-trait GWAS of atherosclerosis-related phenotypes and selected cardiovascular risk factors designed to provide deeper insights into the shared genetic architecture underlying atherosclerosis and enhance power for locus discovery.\u003cstrong\u003e\u0026nbsp;\u003c/strong\u003e\u003c/p\u003e"},{"header":"Results","content":"\u003cdiv id=\"Sec2\" class=\"Section2\"\u003e \u003ch2\u003eCross-trait analysis for atherosclerosis and related risk factors\u003c/h2\u003e \u003cp\u003eWe conducted a multi-stage, multi-trait analysis using GWAS summary statistics collected from 3 different measures of atherosclerosis (CAD, CAC, and CIMT) and 3 risk factors (T2D, LDL-C, SBP, Fig.\u0026nbsp;\u003cspan refid=\"Fig2\" class=\"InternalRef\"\u003e1\u003c/span\u003e, \u003cb\u003eSupplemental Table\u0026nbsp;1\u003c/b\u003e).\u003csup\u003e\u003cspan citationid=\"CR10\" class=\"CitationRef\"\u003e10\u003c/span\u003e,\u003cspan citationid=\"CR12\" class=\"CitationRef\"\u003e12\u003c/span\u003e,\u003cspan citationid=\"CR14\" class=\"CitationRef\"\u003e14\u003c/span\u003e,\u003cspan citationid=\"CR16\" class=\"CitationRef\"\u003e16\u003c/span\u003e,\u003cspan citationid=\"CR17\" class=\"CitationRef\"\u003e17\u003c/span\u003e,\u003cspan citationid=\"CR19\" class=\"CitationRef\"\u003e19\u003c/span\u003e,\u003cspan citationid=\"CR20\" class=\"CitationRef\"\u003e20\u003c/span\u003e\u003c/sup\u003e First, we conducted cross-trait linkage disequilibrium (LD) score regression using LD Score v1.0 (LDSC) to estimate genetic correlation, and shared heritability for each of the 12 pairs of traits.\u003csup\u003e\u003cspan citationid=\"CR27\" class=\"CitationRef\"\u003e27\u003c/span\u003e\u003c/sup\u003e Multi-ancestry GWAS summary statistics were available for all traits except SBP where summary statistics were only available from those of European ancestry at the time of this analysis. Analyses were performed using the 1000 Genomes European LD reference panel.\u003csup\u003e\u003cspan citationid=\"CR28\" class=\"CitationRef\"\u003e28\u003c/span\u003e\u003c/sup\u003e All atherosclerosis and related risk factor traits were significantly (P\u0026thinsp;\u0026lt;\u0026thinsp;0.05) genetically correlated with CAD (Fig.\u0026nbsp;\u003cspan refid=\"Fig4\" class=\"InternalRef\"\u003e2\u003c/span\u003e). Both CAC and CIMT were significantly genetically correlated with T2D and SBP. CAC was also significantly genetically correlated with LDL-C, but CIMT was not. The strongest observed genetic correlation across all traits was between CAC and CAD [\u003cem\u003er\u003c/em\u003e\u003csub\u003e\u003cem\u003eg\u003c/em\u003e\u003c/sub\u003e=0.74\u0026thinsp;\u0026plusmn;\u0026thinsp;0.05 (standard error)], followed by CAD and T2D [\u003cem\u003er\u003c/em\u003e\u003csub\u003e\u003cem\u003eg\u003c/em\u003e\u003c/sub\u003e=0.34\u0026thinsp;\u0026plusmn;\u0026thinsp;0.02]. For studies that had ancestry-specific datasets available, we repeated the analysis using European-only datasets and found the genetic correlations to be similar with only slight variations in the magnitude and precision of the estimates (\u003cb\u003eSupplemental Table\u0026nbsp;2\u003c/b\u003e, Fig.\u0026nbsp;\u003cspan refid=\"Fig4\" class=\"InternalRef\"\u003e2\u003c/span\u003e).\u003c/p\u003e \u003cp\u003e​\u003cem\u003eMulti-trait GWAS\u003c/em\u003e\u003c/p\u003e \u003cp\u003eWe conducted a multi-stage, multi-trait GWAS using N-weighted multivariate genome-wide association meta-analysis (N-GWAMA).\u003csup\u003e\u003cspan citationid=\"CR23\" class=\"CitationRef\"\u003e23\u003c/span\u003e\u003c/sup\u003e N-GWAMA applies the estimates from cross-trait LD-score regression to re-weight test statistics from single-trait GWAS summary statistics, by sample size and estimated heritability, while adjusting for genetic covariance across traits. We conducted 15 multi-trait analyses (Fig.\u0026nbsp;\u003cspan refid=\"Fig2\" class=\"InternalRef\"\u003e1\u003c/span\u003e) across three stages to systematically assess the genetic architecture of atherosclerosis-related traits and risk factors. Stage 1 focused on atherosclerosis traits (CAD, CAC, CIMT) to capture shared genetic associations within clinically relevant disease endpoints. Stage 2 incorporated subclinical traits and selected risk factors (e.g., LDL-C, SBP, T2D) to assess their independent and joint contributions. Stage 3 integrated both approaches, combining atherosclerosis traits with risk factors (e.g., CAD-CAC-SBP) to dissect the genetic pathways underlying disease progression from subclinical to clinical disease. By structuring our analysis in this manner, we aimed to maximize power for variant discovery while mitigating dilution effects from unassociated phenotypes. This approach also enabled us to differentiate variants associated with distinct biological pathways contributing to atherosclerosis development.\u003c/p\u003e \u003cp\u003eWe used Functional Mapping and Annotation of GWAS (FUMA) to annotate the summary results generated from N-GWAMA.\u003csup\u003e\u003cspan citationid=\"CR29\" class=\"CitationRef\"\u003e29\u003c/span\u003e\u003c/sup\u003e We identified 1,177 multi-trait risk loci at a GWAS significance threshold of 5\u0026times;10\u003csup\u003e\u0026minus;\u0026thinsp;8\u003c/sup\u003e, of which 948 met the experiment-wide Bonferroni-corrected significance threshold of 3.3\u0026times;10\u003csup\u003e\u0026minus;\u0026thinsp;9\u003c/sup\u003e (5\u0026times;10\u003csup\u003e\u0026minus;\u0026thinsp;8\u003c/sup\u003e\u0026divide;15) across all stages of analysis (\u003cb\u003eSupplemental Fig.\u0026nbsp;1\u0026ndash;3\u003c/b\u003e). A risk locus was considered novel for atherosclerosis if the lead variant did not meet the genome-wide significance threshold in any of the initial single-trait summary statistics for CAC, CAD, or CIMT and was located more than 500 kb away from a previously reported genome-wide significant variant in the GWAS Catalog.\u003csup\u003e\u003cspan citationid=\"CR30\" class=\"CitationRef\"\u003e30\u003c/span\u003e\u003c/sup\u003e For Stage 2 analyses which were restricted to subclinical atherosclerosis traits, a locus was considered novel only if it had not been reported in the initial CAD summary statistics and was also more than 500 kb away from any genome-wide significant variant for CAD. This approach ensured that all reported loci were novel for all measures of atherosclerosis regardless of analysis stage.\u003c/p\u003e \u003cp\u003eWe identified 173 significant loci during Stage 1 analyses. Most of the significant loci had known associations with at least one atherosclerosis trait (CAD, CAC, or CIMT) (\u003cb\u003eSupplemental Table\u0026nbsp;3\u003c/b\u003e). Only one locus, rs472784 in \u003cem\u003eDLG2\u003c/em\u003e (P\u0026thinsp;=\u0026thinsp;2.0\u0026times;10\u003csup\u003e\u0026minus;\u0026thinsp;9\u003c/sup\u003e), from the CAD-CIMT analysis was novel for atherosclerosis. Similarly, in Stage 2 analyses, 451 experiment-wide significant loci were identified and 210 were novel atherosclerosis loci (\u003cb\u003eSupplemental Table\u0026nbsp;3, Supplemental Table\u0026nbsp;4\u003c/b\u003e). Of these, 10 loci were also novel for the included risk factor. Finally, during Stage 3 analyses, we identified 324 significant loci with 115 novel loci for atherosclerosis. There were 17 significant loci that were also novel for the selected risk factor in the analysis (\u003cb\u003eSupplemental Table\u0026nbsp;3, Supplemental Table\u0026nbsp;5\u003c/b\u003e).\u003c/p\u003e \u003cp\u003eTo identify shared and unique loci across all multi-trait analyses, we merged characterized loci into a single shared genomic locus if the lead variants from different analyses were within 500 kb of each other. For each shared locus, the variant with the most significant P-value across analyses was designated as the shared lead variant. Thus, the 1,177 multi-trait risk loci at GWAS significance threshold of 5\u0026times;10\u003csup\u003e\u0026minus;\u0026thinsp;8\u003c/sup\u003e collapsed into 535 shared loci across all stages of analysis, with 442 containing at least one experiment-wide significant multi-trait locus (Fig.\u0026nbsp;\u003cspan refid=\"Fig6\" class=\"InternalRef\"\u003e3\u003c/span\u003e, \u003cb\u003eSupplemental Table\u0026nbsp;6)\u003c/b\u003e. Most of the 442 significant shared risk loci were known genomic regions that have previously been associated with atherosclerosis, with 195 shared risk loci that were novel for atherosclerosis. There were 25 loci that were also novel for a selected risk factor in the multi-trait analysis. Half of the novel atherosclerosis loci were identified in analyses restricted to subclinical atherosclerosis (101/195\u0026thinsp;=\u0026thinsp;51%). Overall, there were 60 novel atherosclerosis loci associated with CIMT, 27 loci associated with CAC, and 14 novel atherosclerosis loci overlapping with CAC and CIMT. The remaining novel atherosclerosis loci were identified in analyses with CAD (N\u0026thinsp;=\u0026thinsp;94). There were 41 novel atherosclerosis loci shared between CAD and CIMT, 30 shared between CAC and CAD and 23 shared across all atherosclerosis traits (CAD, CAC, and CIMT). For novel atherosclerosis loci, 4 regions overlapped the most frequently across all stages of multi-trait analyses (nearest gene: \u003cem\u003eBNC2\u003c/em\u003e, \u003cem\u003eGPATCH2\u003c/em\u003e, \u003cem\u003eINSR\u003c/em\u003e, \u003cem\u003eJAZF1\u003c/em\u003e).\u003c/p\u003e \u003cp\u003eMost of the experiment-wide significant loci were identified in multi-trait analyses which included SBP (Fig.\u0026nbsp;\u003cspan refid=\"Fig6\" class=\"InternalRef\"\u003e3\u003c/span\u003e). Distinct groups of shared genomic regions were identified across atherosclerosis traits. There were 50 shared loci shared across SBP, CIMT, and CAD, 58 shared loci that included just CIMT-SBP, 37 shared across all atherosclerosis traits and SBP and 35 shared loci that included CAC-CAD-SBP. Similar patterns were noted with both remaining risk factor traits with various trait combinations identifying important pleiotropic genomic regions for atherosclerosis. Six shared loci that were shared across all traits in the analysis, all with known associations with atherosclerosis (nearest genes: \u003cem\u003eMAT2A, IRS1, STAG1, PVRL2, OPRL1, ARVCF\u003c/em\u003e).\u003c/p\u003e \u003c/div\u003e \u003cdiv id=\"Sec3\" class=\"Section2\"\u003e \u003ch2\u003eTrait-trait and Trait- eQTL colocalization\u003c/h2\u003e \u003cp\u003eMulti-trait colocalization was conducted using HyPrColoc, a Bayesian divisive clustering algorithm designed to identify shared causal signals within a genomic region.\u003csup\u003e\u003cspan citationid=\"CR22\" class=\"CitationRef\"\u003e22\u003c/span\u003e\u003c/sup\u003e. Evidence for colocalization was determined based on the default variant-specific regional and alignment priors (P\u003csub\u003e\u003cem\u003eR\u003c/em\u003e\u003c/sub\u003e\u0026lowast;=P\u003csub\u003e\u003cem\u003eA\u003c/em\u003e\u003c/sub\u003e\u003cem\u003e\u0026lowast;\u003c/em\u003e=0.5), with colocalization identified when P\u003csub\u003e\u003cem\u003eR\u003c/em\u003e\u003c/sub\u003eP\u003csub\u003e\u003cem\u003eA\u003c/em\u003e\u003c/sub\u003e\u0026ge;0.25. Strong evidence of colocalization across traits was defined as P\u003csub\u003eR\u003c/sub\u003eP\u003csub\u003eA\u003c/sub\u003e \u0026ge; 0.80. Overall, 164 significant shared risk loci identified in the multi-trait analysis also had evidence of multi-trait colocalization with a measure of atherosclerosis in HyPrColoc. Overall, multi-trait GWAS analyses that included SBP also colocalized the most frequently with the respective atherosclerosis traits (N\u0026thinsp;=\u0026thinsp;164), followed by LDL-C (N\u0026thinsp;=\u0026thinsp;93), and T2D (N\u0026thinsp;=\u0026thinsp;36). There were 25 novel atherosclerosis loci with evidence of colocalization \u003cb\u003e(Supplemental Table\u0026nbsp;7).\u003c/b\u003e Of the 25 shared novel atherosclerosis loci with evidence of colocalization, 7 analyses strongly colocalized with P\u003csub\u003eR\u003c/sub\u003eP\u003csub\u003eA\u003c/sub\u003e \u0026ge;0.80 (nearest gene: \u003cem\u003eBNC2, SCAI, TSC22D2, SRRM1, ABCB11, PRRX2\u003c/em\u003e [Figure \u003cspan refid=\"Fig8\" class=\"InternalRef\"\u003e4\u003c/span\u003e]).\u003c/p\u003e \u003cp\u003eFinally, for loci that colocalized in the trait-trait analysis, we performed trait- expression quantitative loci (eQTL) colocalization using eQTL data from GTEX v8 using the COLOC package in R.\u003csup\u003e\u003cspan citationid=\"CR31\" class=\"CitationRef\"\u003e31\u003c/span\u003e,32 33\u003c/sup\u003e Using summary statistics from the multi-trait GWAS for corresponding trait-trait pairwise colocalization, we focused our analysis on a subset of GTEx v8 tissues selected based on their biological relevance to the studied traits (\u003cb\u003eSupplemental Table\u0026nbsp;8\u003c/b\u003e). Evidence for colocalization was defined as a posterior probability of a shared causal variant (PP\u003csub\u003eH4\u003c/sub\u003e)\u0026thinsp;\u0026ge;\u0026thinsp;0.50 and a conditional posterior probability (PP\u003csub\u003ec\u003c/sub\u003e=PP\u003csub\u003eH4\u003c/sub\u003e/(PP\u003csub\u003eH3\u003c/sub\u003e+PP\u003csub\u003eH4)\u003c/sub\u003e)\u0026thinsp;\u0026ge;\u0026thinsp;0.80. Strong evidence for colocalization was defined as (PP\u003csub\u003eH4\u003c/sub\u003e)\u0026thinsp;\u0026ge;\u0026thinsp;0.80 and a conditional posterior probability (PP\u003csub\u003eH4\u003c/sub\u003e/(PP\u003csub\u003eH3\u003c/sub\u003e+PP\u003csub\u003eH4\u003c/sub\u003e))\u0026thinsp;\u0026ge;\u0026thinsp;0.80.\u003c/p\u003e \u003cp\u003eWe found evidence of colocalization (PP\u003csub\u003eH4\u003c/sub\u003e \u0026ge;0.50 and PP\u003csub\u003ec\u003c/sub\u003e\u0026ge;0.80) with eQTL data from GTEx with novel atherosclerosis loci. Evidence of colocalization was identified most frequently with eQTLs in adipose tissue and arterial tissue of the Aorta and Tibia (N\u0026thinsp;=\u0026thinsp;132, 114, and 112, respectively). There were 18 significant novel atherosclerosis loci with evidence of colocalization with eQTL data from GTEx that had evidence of trait-trait colocalization as well (Table\u0026nbsp;\u003cspan refid=\"Tab1\" class=\"InternalRef\"\u003e1\u003c/span\u003e). We identified 5 genes with strong evidence for colocalization (P\u003csub\u003eR\u003c/sub\u003eP\u003csub\u003eA\u003c/sub\u003e \u0026ge;0.80, PP\u003csub\u003eH4\u003c/sub\u003e \u0026ge;0.80, and PP\u003csub\u003ec\u003c/sub\u003e \u0026ge;0.80) in trait-eQTL and in trait-trait colocalization analysis (\u003cem\u003ePRRX2\u003c/em\u003e, \u003cem\u003eBNC2\u003c/em\u003e, \u003cem\u003eCLIC4\u003c/em\u003e, \u003cem\u003eSCAI\u003c/em\u003e, and \u003cem\u003ePPP6C).\u003c/em\u003e SBP and CIMT colocalized with expression of \u003cem\u003ePRRX2, SCAI\u003c/em\u003e, and \u003cem\u003ePPP6C\u003c/em\u003e in arterial tissues (Fig.\u0026nbsp;\u003cspan refid=\"Fig10\" class=\"InternalRef\"\u003e5\u003c/span\u003e). These findings suggest a potential mechanistic link between vascular gene regulation and atherosclerosis traits. The expression of \u003cem\u003eBNC2\u003c/em\u003e in whole blood colocalized with CAC, CAD, CIMT, and LDL-C (Fig.\u0026nbsp;\u003cspan refid=\"Fig12\" class=\"InternalRef\"\u003e6\u003c/span\u003e), while the expression of \u003cem\u003eCLIC4\u003c/em\u003e in visceral omentum adipose tissue and left ventricle heart tissue colocalized with CAD, CAC, CIMT, and SBP (Fig.\u0026nbsp;\u003cspan refid=\"Fig14\" class=\"InternalRef\"\u003e7\u003c/span\u003e).\u003c/p\u003e \u003cp\u003e \u003cdiv class=\"gridtable\"\u003e\u003ctable float=\"Yes\" id=\"Tab1\" border=\"1\"\u003e \u003ccaption language=\"En\"\u003e \u003cdiv class=\"CaptionNumber\"\u003eTable 1\u003c/div\u003e \u003cdiv class=\"CaptionContent\"\u003e \u003cp\u003eNovel atherosclerosis risk loci that colocalized with tissue-specific gene expression levels.\u003c/p\u003e \u003c/div\u003e \u003c/caption\u003e \u003ccolgroup cols=\"7\"\u003e \u003cdiv align=\"left\" class=\"colspec\" colname=\"c1\" colnum=\"1\"\u003e\u003c/div\u003e \u003cdiv align=\"left\" class=\"colspec\" colname=\"c2\" colnum=\"2\"\u003e\u003c/div\u003e \u003cdiv align=\"left\" class=\"colspec\" colname=\"c3\" colnum=\"3\"\u003e\u003c/div\u003e \u003cdiv align=\"left\" class=\"colspec\" colname=\"c4\" colnum=\"4\"\u003e\u003c/div\u003e \u003cdiv align=\"left\" class=\"colspec\" colname=\"c5\" colnum=\"5\"\u003e\u003c/div\u003e \u003cdiv align=\"left\" class=\"colspec\" colname=\"c6\" colnum=\"6\"\u003e\u003c/div\u003e \u003cdiv align=\"left\" class=\"colspec\" colname=\"c7\" colnum=\"7\"\u003e\u003c/div\u003e \u003cthead\u003e \u003ctr\u003e \u003cth align=\"left\" colname=\"c1\"\u003e \u003cp\u003eShared Locus ID\u003csup\u003e\u003cspan citationid=\"CR1\" class=\"CitationRef\"\u003e1\u003c/span\u003e\u003c/sup\u003e\u003c/p\u003e \u003c/th\u003e \u003cth align=\"left\" colname=\"c2\"\u003e \u003cp\u003eTraits in\u003c/p\u003e \u003cp\u003eMulti-trait GWAS\u003c/p\u003e \u003c/th\u003e \u003cth align=\"left\" colname=\"c3\"\u003e \u003cp\u003eGene Symbol\u003csup\u003e\u003cspan citationid=\"CR2\" class=\"CitationRef\"\u003e2\u003c/span\u003e\u003c/sup\u003e\u003c/p\u003e \u003c/th\u003e \u003cth align=\"left\" colname=\"c4\"\u003e \u003cp\u003eTissue Type\u003csup\u003e\u003cspan citationid=\"CR3\" class=\"CitationRef\"\u003e3\u003c/span\u003e\u003c/sup\u003e\u003c/p\u003e \u003c/th\u003e \u003cth align=\"left\" colname=\"c5\"\u003e \u003cp\u003ePosterior Probability (H4)\u003csup\u003e\u003cspan citationid=\"CR4\" class=\"CitationRef\"\u003e4\u003c/span\u003e\u003c/sup\u003e\u003c/p\u003e \u003c/th\u003e \u003cth align=\"left\" colname=\"c6\"\u003e \u003cp\u003eConditional\u003c/p\u003e \u003cp\u003ePosterior\u003c/p\u003e \u003cp\u003eProbability\u003csup\u003e\u003cspan citationid=\"CR5\" class=\"CitationRef\"\u003e5\u003c/span\u003e\u003c/sup\u003e\u003c/p\u003e \u003c/th\u003e \u003cth align=\"left\" colname=\"c7\"\u003e \u003cp\u003eCandidate\u003c/p\u003e \u003cp\u003eCausal\u003c/p\u003e \u003cp\u003eSNP\u003c/p\u003e \u003c/th\u003e \u003c/tr\u003e \u003c/thead\u003e \u003ctbody\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e \u003cp\u003e276\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c2\"\u003e \u003cp\u003eCAC, CAD, LDL-C\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c3\"\u003e \u003cp\u003e\u003cem\u003eBNC2\u003c/em\u003e\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c4\"\u003e \u003cp\u003eWhole Blood\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c5\"\u003e \u003cp\u003e0.98\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c6\"\u003e \u003cp\u003e0.98\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c7\"\u003e \u003cp\u003ers28498684\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e \u003cp\u003e276\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c2\"\u003e \u003cp\u003eCAD, CIMT, LDL-C\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c3\"\u003e \u003cp\u003e\u003cem\u003eBNC2\u003c/em\u003e\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c4\"\u003e \u003cp\u003eWhole Blood\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c5\"\u003e \u003cp\u003e0.98\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c6\"\u003e \u003cp\u003e0.98\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c7\"\u003e \u003cp\u003ers28498684\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e \u003cp\u003e6\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c2\"\u003e \u003cp\u003eSBP, CAC, CAD\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c3\"\u003e \u003cp\u003e\u003cem\u003eCLIC4\u003c/em\u003e\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c4\"\u003e \u003cp\u003eLeft Ventricle\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c5\"\u003e \u003cp\u003e0.93\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c6\"\u003e \u003cp\u003e0.94\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c7\"\u003e \u003cp\u003ers4366267\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e\u0026nbsp;\u003c/td\u003e \u003ctd align=\"left\" colname=\"c2\"\u003e\u0026nbsp;\u003c/td\u003e \u003ctd align=\"left\" colname=\"c3\"\u003e\u0026nbsp;\u003c/td\u003e \u003ctd align=\"left\" colname=\"c4\"\u003e \u003cp\u003eVisceral Adipose\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c5\"\u003e \u003cp\u003e0.96\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c6\"\u003e \u003cp\u003e0.97\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c7\"\u003e \u003cp\u003ers4366267\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e \u003cp\u003e6\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c2\"\u003e \u003cp\u003eSBP, CIMT, CAD\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c3\"\u003e \u003cp\u003e\u003cem\u003eCLIC4\u003c/em\u003e\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c4\"\u003e \u003cp\u003eLeft Ventricle\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c5\"\u003e \u003cp\u003e0.92\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c6\"\u003e \u003cp\u003e0.94\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c7\"\u003e \u003cp\u003ers72654647\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e\u0026nbsp;\u003c/td\u003e \u003ctd align=\"left\" colname=\"c2\"\u003e\u0026nbsp;\u003c/td\u003e \u003ctd align=\"left\" colname=\"c3\"\u003e\u0026nbsp;\u003c/td\u003e \u003ctd align=\"left\" colname=\"c4\"\u003e \u003cp\u003eVisceral Adipose\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c5\"\u003e \u003cp\u003e0.97\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c6\"\u003e \u003cp\u003e0.97\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c7\"\u003e \u003cp\u003ers6686889\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e \u003cp\u003e287\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c2\"\u003e \u003cp\u003eSBP, CIMT\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c3\"\u003e \u003cp\u003e\u003cem\u003ePPP6C\u003c/em\u003e\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c4\"\u003e \u003cp\u003eTibial Artery\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c5\"\u003e \u003cp\u003e0.83\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c6\"\u003e \u003cp\u003e0.85\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c7\"\u003e \u003cp\u003ers72765265\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e\u0026nbsp;\u003c/td\u003e \u003ctd align=\"left\" colname=\"c2\"\u003e\u0026nbsp;\u003c/td\u003e \u003ctd align=\"left\" colname=\"c3\"\u003e \u003cp\u003e\u003cem\u003eSCAI\u003c/em\u003e\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c4\"\u003e \u003cp\u003eAorta Artery\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c5\"\u003e \u003cp\u003e0.95\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c6\"\u003e \u003cp\u003e0.95\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c7\"\u003e \u003cp\u003ers11793512\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e\u0026nbsp;\u003c/td\u003e \u003ctd align=\"left\" colname=\"c2\"\u003e\u0026nbsp;\u003c/td\u003e \u003ctd align=\"left\" colname=\"c3\"\u003e\u0026nbsp;\u003c/td\u003e \u003ctd align=\"left\" colname=\"c4\"\u003e \u003cp\u003eTibial Artery\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c5\"\u003e \u003cp\u003e0.93\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c6\"\u003e \u003cp\u003e0.93\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c7\"\u003e \u003cp\u003ers11793475\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e \u003cp\u003e289**\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c2\"\u003e \u003cp\u003eSBP, CIMT\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c3\"\u003e \u003cp\u003e\u003cem\u003ePRRX2\u003c/em\u003e\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c4\"\u003e \u003cp\u003eAorta Artery\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c5\"\u003e \u003cp\u003e0.98\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c6\"\u003e \u003cp\u003e0.98\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c7\"\u003e \u003cp\u003ers11788582\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e\u0026nbsp;\u003c/td\u003e \u003ctd align=\"left\" colname=\"c2\"\u003e\u0026nbsp;\u003c/td\u003e \u003ctd align=\"left\" colname=\"c3\"\u003e\u0026nbsp;\u003c/td\u003e \u003ctd align=\"left\" colname=\"c4\"\u003e \u003cp\u003eTibial Artery\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c5\"\u003e \u003cp\u003e0.98\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c6\"\u003e \u003cp\u003e0.98\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c7\"\u003e \u003cp\u003ers11788582\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e\u0026nbsp;\u003c/td\u003e \u003ctd align=\"left\" colname=\"c2\"\u003e\u0026nbsp;\u003c/td\u003e \u003ctd align=\"left\" colname=\"c3\"\u003e\u0026nbsp;\u003c/td\u003e \u003ctd align=\"left\" colname=\"c4\"\u003e \u003cp\u003eLeft Ventricle\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c5\"\u003e \u003cp\u003e0.97\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c6\"\u003e \u003cp\u003e0.98\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c7\"\u003e \u003cp\u003ers920659\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e\u0026nbsp;\u003c/td\u003e \u003ctd align=\"left\" colname=\"c2\"\u003e\u0026nbsp;\u003c/td\u003e \u003ctd align=\"left\" colname=\"c3\"\u003e\u0026nbsp;\u003c/td\u003e \u003ctd align=\"left\" colname=\"c4\"\u003e \u003cp\u003eSubcutaneous Adipose\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c5\"\u003e \u003cp\u003e0.80\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c6\"\u003e \u003cp\u003e0.84\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c7\"\u003e \u003cp\u003ers920659\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e\u0026nbsp;\u003c/td\u003e \u003ctd align=\"left\" colname=\"c2\"\u003e\u0026nbsp;\u003c/td\u003e \u003ctd align=\"left\" colname=\"c3\"\u003e\u0026nbsp;\u003c/td\u003e \u003ctd align=\"left\" colname=\"c4\"\u003e \u003cp\u003eVisceral Omentum Adipose\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c5\"\u003e \u003cp\u003e0.98\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c6\"\u003e \u003cp\u003e0.98\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c7\"\u003e \u003cp\u003ers13299355\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e\u0026nbsp;\u003c/td\u003e \u003ctd align=\"left\" colname=\"c2\"\u003e\u0026nbsp;\u003c/td\u003e \u003ctd align=\"left\" colname=\"c3\"\u003e\u0026nbsp;\u003c/td\u003e \u003ctd align=\"left\" colname=\"c4\"\u003e \u003cp\u003eCultured fibroblasts Cells\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c5\"\u003e \u003cp\u003e0.98\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c6\"\u003e \u003cp\u003e0.98\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c7\"\u003e \u003cp\u003ers59878076\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colspan=\"7\" nameend=\"c7\" namest=\"c1\"\u003e \u003cp\u003eAbbreviations: CAC, coronary artery calcification; CAD, coronary artery disease; CIMT. carotid intima- media thickness; GWAS, genome-wide association study; LDL-C, low- density lipoprotein cholesterol; SBP, systolic blood pressure, SNP, single nucleotide polymorphism\u003c/p\u003e \u003cp\u003e\u003csup\u003e\u003cspan citationid=\"CR1\" class=\"CitationRef\"\u003e1\u003c/span\u003e\u003c/sup\u003e Shared Locus ID is a unique identifier for loci shared across multiple traits, defined as SNPs within 500 kb of each other. ** indicates novel for both atherosclerosis and associated risk factors.\u003c/p\u003e \u003cp\u003e\u003csup\u003e\u003cspan citationid=\"CR2\" class=\"CitationRef\"\u003e2\u003c/span\u003e\u003c/sup\u003e HGNC Gene Symbol represents the gene whose eQTL colocalized with the genomic locus, suggesting a regulatory role.\u003c/p\u003e \u003cp\u003e\u003csup\u003e\u003cspan citationid=\"CR3\" class=\"CitationRef\"\u003e3\u003c/span\u003e\u003c/sup\u003e Tissue indicates where the colocalized signal was detected, based on GTEx or other tissue-specific datasets.\u003c/p\u003e \u003cp\u003e\u003csup\u003e\u003cspan citationid=\"CR4\" class=\"CitationRef\"\u003e4\u003c/span\u003e\u003c/sup\u003e Posterior Probability (H4) represents the probability that the signal is shared between traits under the H4 model of colocalization.\u003c/p\u003e \u003cp\u003e\u003csup\u003e\u003cspan citationid=\"CR5\" class=\"CitationRef\"\u003e5\u003c/span\u003e\u003c/sup\u003e Conditional Posterior Probability reflects the posterior probability after conditioning on other signals in the locus.\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003c/tbody\u003e \u003c/colgroup\u003e \u003c/table\u003e\u003c/div\u003e \u003c/p\u003e \u003c/div\u003e"},{"header":"Discussion","content":"\u003cp\u003eHere we highlight the value of using existing, publicly available data to conduct a multi-stage, multi-trait analysis of related complex traits to understand shared genetic architecture for atherosclerosis and select risk factor traits. Furthermore, this approach allowed us to identify novel pleiotropic susceptibility loci for atherosclerosis. Specifically, we identified 195 shared loci that were novel for atherosclerosis and met our experiment-wide significance threshold, all of which underscore specific underlying pathways linked to subclinical atherosclerosis and an associated risk factor. Multi-trait colocalization further confirmed shared causal signals between atherosclerosis and selected risk factors at 25 novel atherosclerosis loci. Additionally, we integrated gene expression and eQTL data from GTEx to refine the multi-trait signals and identify functional insights for candidate genes involved in the underlying atherosclerosis pathogenesis.\u003c/p\u003e \u003cp\u003eOur study underscores the importance of leveraging multi-trait analysis for complex phenotypes like CAD and measures of subclinical atherosclerosis. It is well-established in the literature that CAD shares genomic risk factors with related biological traits and diseases, such as SBP and T2D.\u003csup\u003e\u003cspan citationid=\"CR9\" class=\"CitationRef\"\u003e9\u003c/span\u003e,\u003cspan citationid=\"CR12\" class=\"CitationRef\"\u003e12\u003c/span\u003e,\u003cspan citationid=\"CR13\" class=\"CitationRef\"\u003e13\u003c/span\u003e,\u003cspan citationid=\"CR24\" class=\"CitationRef\"\u003e24\u003c/span\u003e\u003c/sup\u003e While single-trait GWAS continues to identify novel loci, additional resources and follow-up analyses are needed to contextualize newly discovered genetic variants. Our analysis identified 195 novel loci for atherosclerosis, 94 of which were associated with CAD, subclinical atherosclerosis, and their respective risk factors, providing further evidence for their potential roles in clinical disease. Additionally, 101 loci were identified only in analyses with CIMT or CAC and respective risk factors, suggesting a specific role in the early stages of atherosclerosis development. These findings highlight the complementary utility of subclinical traits like CIMT and CAC in uncovering potential novel genetic pathways that precede clinical disease.\u003csup\u003e\u003cspan citationid=\"CR19\" class=\"CitationRef\"\u003e19\u003c/span\u003e\u003c/sup\u003e\u003c/p\u003e \u003cp\u003eImportantly, we identified potential regulatory roles involved in the development of atherosclerosis by integrating tissue-specific gene expression data with novel atherosclerosis loci. We identified colocalization between the multi-trait GWAS and tissue-specific gene expression levels at 4 novel loci, involving 5 genes, including \u003cem\u003eSCAI\u003c/em\u003e, \u003cem\u003ePRRX2\u003c/em\u003e, and \u003cem\u003eCLIC4\u003c/em\u003e. \u003cem\u003eSCAI\u003c/em\u003e expression in arterial tissue colocalized with CIMT, and SBP. SCAI is a negative regulator of Rho protein activation, particularly in the RhoA/DIAPH1 pathway.\u003csup\u003e\u003cspan citationid=\"CR34\" class=\"CitationRef\"\u003e34\u003c/span\u003e\u003c/sup\u003e Prior studies indicate that \u003cem\u003eDIAPH1\u003c/em\u003e knockout in mice attenuates atherosclerosis progression, and downregulation of \u003cem\u003eDIAPH1\u003c/em\u003e expression has been observed in ischemic stroke patients.\u003csup\u003e\u003cspan additionalcitationids=\"CR36\" citationid=\"CR35\" class=\"CitationRef\"\u003e35\u003c/span\u003e\u0026ndash;\u003cspan citationid=\"CR37\" class=\"CitationRef\"\u003e37\u003c/span\u003e\u003c/sup\u003e These findings suggest that SCAI may influence both structural and functional aspects of vascular biology, warranting further investigation as a potential target for atherosclerosis research. Similarly, CIMT and SBP also colocalized with \u003cem\u003ePRRX2\u003c/em\u003e expression in central and peripheral arterial tissues. PRRX2 is a transcription factor involved in vascular smooth muscle cell differentiation and migration, with established roles in cardiovascular development during embryogenesis.\u003csup\u003e\u003cspan citationid=\"CR38\" class=\"CitationRef\"\u003e38\u003c/span\u003e\u003c/sup\u003e A recent study linked the upregulation of \u003cem\u003ePRRX2\u003c/em\u003e signaling to cardiac remodeling post-myocardial infarction in a mouse model, indicating its potential involvement in SBP and CIMT through arterial vascular smooth muscle cell proliferation or remodeling.\u003c/p\u003e \u003cp\u003eWe also identified a shared causal signal involving \u003cem\u003eCLIC4\u003c/em\u003e expression in the left ventricle and visceral omentum adipose tissue, that was associated with multi-trait signals including CAD-IMT-SBP and CAD-CAC-SBP. \u003cem\u003eCLIC4\u003c/em\u003e is implicated in apoptosis and inflammation processes that are critical to atherosclerosis development.\u003csup\u003e\u003cspan additionalcitationids=\"CR40\" citationid=\"CR39\" class=\"CitationRef\"\u003e39\u003c/span\u003e\u0026ndash;\u003cspan citationid=\"CR41\" class=\"CitationRef\"\u003e41\u003c/span\u003e\u003c/sup\u003e Its significant colocalization with CAC, CAD, CIMT, and SBP suggests it may be a central regulator of cardiovascular and metabolic health. Emerging research using in vitro atherosclerosis cell models highlights CLIC4's critical role in endothelial cell function and its potential as a therapeutic target for atherosclerosis.\u003csup\u003e\u003cspan additionalcitationids=\"CR40 CR41\" citationid=\"CR39\" class=\"CitationRef\"\u003e39\u003c/span\u003e\u0026ndash;\u003cspan citationid=\"CR42\" class=\"CitationRef\"\u003e42\u003c/span\u003e\u003c/sup\u003e Future studies should explore the mechanistic pathways of \u003cem\u003eCLIC4\u003c/em\u003e in immune and metabolic regulation and its therapeutic potential for atherosclerosis and hypertension.\u003c/p\u003e \u003cp\u003eOur study represents the first multi-trait GWAS to integrate both clinical and subclinical atherosclerosis, leveraging summary statistics from CAD, CAC, CIMT, SBP, LDL-C and T2D to enhance power for detecting novel loci associated with both early and late stages of atherosclerosis progression. This approach provides valuable insights into genetic mechanisms with potential implications for early prevention. By incorporating the largest available GWAS datasets for these traits, we offer a comprehensive perspective on their shared genetic architecture. The multi-trait framework enables the identification of pleiotropic genetic effects, refining risk loci with greater precision and uncovering shared pathways that contribute to atherosclerosis. Additionally, by integrating colocalization and functional genomic analyses, our study provides deeper biological insights linking genetic variants to gene expression and potential causal mechanisms. Addressing heterogeneity in disease progression through the inclusion of both subclinical and clinical phenotypes, our approach captures a broader spectrum of atherosclerosis development, revealing novel insights into genetic factors contributing to both early and late-stage disease, paving the way for potential early intervention and personalized prevention strategies.\u003c/p\u003e \u003cp\u003eNevertheless, this our study is not without limitations. While the GWAS summary statistics include multiple population groups, non-European populations are underrepresented. Cross-population genetic correlations required the use of population-specific reference panels for genetic covariance calculations. Our sensitivity analyses demonstrated similar genetic correlations between cross-population and primarily European GWAS studies, emphasizing European ancestry representation in our dataset. Future studies would benefit from using more diverse LD reference panels and genetic correlation methods that account for multiple genetically inferred genetic ancestral groups. Additionally, the power of our multi-trait colocalization analysis was limited by differences in LD patterns, likely stemming from varying ancestry distributions in the summary statistics. Furthermore, the GWAS for subclinical traits had smaller sample sizes and larger standard errors compared to the GWAS for CAD, LDL-C, T2D, and SBP datasets. Consequently, these disparities in data quality potentially affected the multi-trait colocalization analysis, highlighting the need for larger and more ancestrally balanced datasets for future studies. Finally, we recognize that our analyses was limited to a select number of biological risk factors and may lead to an over-identification of specific biological pathways driven by SBP, T2D, and LDL-C while underrepresenting others. We selected traits based on the strength of their known relationships with not only CAD, but CAC, and CIMT, and prioritized large, well-conducted GWAS. Future studies could benefit from evaluating additional biological risk factors or disease endpoints with subclinical atherosclerosis measures to emphasize additional pathways to atherosclerosis.\u003c/p\u003e \u003cp\u003eIn summary, our analyses identified novel loci and pathways involved in the development of atherosclerosis, underscoring the importance of subclinical traits like CIMT and CAC in uncovering early-stage mechanisms and the critical role of SBP in CAD development. Multi-trait integration revealed shared causal signals tied to vascular remodeling, inflammation, and metabolic regulation, implicating genes such as \u003cem\u003eSCAI\u003c/em\u003e, \u003cem\u003ePRRX2\u003c/em\u003e, and \u003cem\u003eCLIC4\u003c/em\u003e as promising candidates for further research. These findings highlight the value of combining subclinical and clinical traits from publicly available GWAS data to bridge early disease processes with clinical outcomes. Future studies should replicate these discoveries in diverse populations, address ancestry-related gaps, and leverage larger datasets to enhance discovery. Nevertheless, this integrative novel approach holds promise for advancing our understanding of CAD pathogenesis and identifying novel therapeutic targets.\u003c/p\u003e"},{"header":"Methods","content":"\u003cp\u003eWe conducted a multi-stage, multi-trait analysis using publicly available GWAS summary statistics for clinical and subclinical atherosclerosis and select cardiometabolic risk factors. We leveraged the largest available CAD, CAC, and CIMT GWAS studies, which included cross-population and European ancestry GWAS, and conducted a multi-stage approach. We conducted multi-trait colocalization analysis across traits and with GTEX v8 (Fig.\u0026nbsp;\u003cspan refid=\"Fig2\" class=\"InternalRef\"\u003e1\u003c/span\u003e).\u003c/p\u003e\n\u003ch3\u003eGenetic Association Studies\u003c/h3\u003e\n\u003cp\u003ePublished GWAS summary statistics from 3 different GWAS on atherosclerosis traits (CAD, CAC, and CIMT) and 3 biological cardiometabolic risk factors (T2D, LDL-C, SBP) were collected.\u003csup\u003e\u003cspan citationid=\"CR11\" class=\"CitationRef\"\u003e11\u003c/span\u003e,\u003cspan citationid=\"CR12\" class=\"CitationRef\"\u003e12\u003c/span\u003e,\u003cspan citationid=\"CR14\" class=\"CitationRef\"\u003e14\u003c/span\u003e,\u003cspan citationid=\"CR16\" class=\"CitationRef\"\u003e16\u003c/span\u003e,\u003cspan citationid=\"CR19\" class=\"CitationRef\"\u003e19\u003c/span\u003e\u003c/sup\u003e We selected traits for inclusion in the multi-trait analysis based on their biological relevance to atherosclerosis, the strength of their relationships with atherosclerosis, and the availability of large, high-quality GWAS datasets. To capture complementary aspects of subclinical atherosclerosis across vascular beds, we included CIMT and CAC. Information about each GWAS is available in \u003cb\u003eSupplemental Table\u0026nbsp;1\u003c/b\u003e, and further details regarding the outcome measure and methods specific to each GWAS are included in the respective publications. All GWAS included well-imputed (r\u003csup\u003e2\u003c/sup\u003e\u0026thinsp;\u0026gt;\u0026thinsp;0.3) low-frequency or common variants (minor allele frequency\u0026thinsp;\u0026gt;\u0026thinsp;1%) using either the 1000 Genomes Project, Haplotype Reference Consortium or TOPMed reference panel.\u003csup\u003e\u003cspan additionalcitationids=\"CR44\" citationid=\"CR43\" class=\"CitationRef\"\u003e43\u003c/span\u003e\u0026ndash;\u003cspan citationid=\"CR45\" class=\"CitationRef\"\u003e45\u003c/span\u003e\u003c/sup\u003e All GWAS consisted of cross-population meta-analyses, with a majority of individuals representing individuals of European ancestry, except for the exception of the SBP GWAS. The SBP GWAS available at the time of this analysis was solely conducted in individuals with European ancestry.\u003csup\u003e\u003cspan citationid=\"CR12\" class=\"CitationRef\"\u003e12\u003c/span\u003e\u003c/sup\u003e\u003c/p\u003e\n\u003ch3\u003eGenetic correlation\u003c/h3\u003e\n\u003cp\u003eCross-trait LD score regression using LD Score v1.0 (LDSC) was used to estimate the sample overlap, genetic correlation, and shared heritability for each trait pair. LDSC evaluates genetic correlation and heritability from GWAS summary statistics using a linkage disequilibrium (LD) reference panel. We conducted the analysis using a European LD reference panel made available from the 1000 Genomes Project.\u003csup\u003e\u003cspan citationid=\"CR27\" class=\"CitationRef\"\u003e27\u003c/span\u003e,\u003cspan citationid=\"CR28\" class=\"CitationRef\"\u003e28\u003c/span\u003e,\u003cspan citationid=\"CR46\" class=\"CitationRef\"\u003e46\u003c/span\u003e\u003c/sup\u003e As a sensitivity analysis, we repeated cross-trait LD score regression for studies that also had summary statistics available that were restricted to European populations.\u003c/p\u003e \u003cdiv id=\"Sec8\" class=\"Section2\"\u003e \u003ch2\u003eMulti-trait analysis\u003c/h2\u003e \u003cp\u003eWe conducted a multi-stage, multi-trait analysis using N-GWAMA. N-GWAMA applies the estimates from cross-trait LD-score regression to re-weight test statistics from single-trait GWAS studies, by sample size and estimated heritability, while adjusting for genetic covariance across traits. This method enhances the detection of loci with pleiotropic associations by leveraging shared genetic architecture across traits. The newly weighted test statistics are then summated and standardized to produce a new test statistic, from which a new p-value can be calculated for each genetic variant.\u003csup\u003e\u003cspan citationid=\"CR27\" class=\"CitationRef\"\u003e27\u003c/span\u003e,\u003cspan citationid=\"CR46\" class=\"CitationRef\"\u003e46\u003c/span\u003e\u003c/sup\u003e We conducted 15 multi-trait analyses using various trait combinations (Fig.\u0026nbsp;\u003cspan refid=\"Fig2\" class=\"InternalRef\"\u003e1\u003c/span\u003e) to identify which combination of traits led to the largest increase in associated variants with atherosclerosis and mitigate any potential power loss from the inclusion of unassociated phenotypes. Stage 1 analyses were restricted to clinical (CAD) and subclinical (CAC and CIMT) atherosclerosis traits. Stage 2 analyses included subclinical atherosclerosis traits with associated risk factors (SBP, LDL-C, T2D). Stage 3 analyses reexamined Stage 2 analyses with the inclusion of CAD (Fig.\u0026nbsp;\u003cspan refid=\"Fig2\" class=\"InternalRef\"\u003e1\u003c/span\u003e).\u003c/p\u003e \u003c/div\u003e\n\u003ch3\u003eFunctional mapping and annotation\u003c/h3\u003e\n\u003cp\u003eWe used Functional Mapping and Annotation of GWAS (FUMA) to annotate the summary results generated from N-GWAMA.\u003csup\u003e\u003cspan citationid=\"CR29\" class=\"CitationRef\"\u003e29\u003c/span\u003e\u003c/sup\u003e First, variants that met the GWAS significance P-value threshold (P\u0026thinsp;\u0026lt;\u0026thinsp;5\u0026times;10\u003csup\u003e\u0026minus;\u0026thinsp;8\u003c/sup\u003e) were initially clumped (r\u003csup\u003e2\u003c/sup\u003e\u0026thinsp;\u0026lt;\u0026thinsp;0.6) using a non-ancestry specific LD reference panel for 1000 Genome Phase 3 to create a genomic risk locus. Within the locus, the variants were clumped again (r\u003csup\u003e2\u003c/sup\u003e\u0026thinsp;\u0026lt;\u0026thinsp;0.1) to identify independent signals. Lead variants were identified as a subset of the independent significant variants with the strongest signal, while other variants were considered secondary signals. If a lead variant was within 250 kb of another lead variant, the loci were combined. Therefore, a genomic risk locus could contain multiple lead variants. Within FUMA, ANNOVAR was used to map variants to genes by position and annotate variants according to predicted consequence and location.\u003csup\u003e\u003cspan citationid=\"CR47\" class=\"CitationRef\"\u003e47\u003c/span\u003e\u003c/sup\u003e\u003c/p\u003e \u003cp\u003eWe reported a genomic risk locus if the lead variant from FUMA had P\u0026thinsp;\u0026lt;\u0026thinsp;0.05 in the included single-trait GWAS summary statistics. A genomic risk locus was considered novel for atherosclerosis and the associated risk factor if the lead variant did not reach the GWAS significance P-value threshold in any of the included single-trait files. A genomic risk locus was considered novel for atherosclerosis if the lead variant did not meet the GWAS genome-wide significance threshold in the single-trait atherosclerosis summary statistics and was more than was more than \u0026plusmn;\u0026thinsp;500 kb away from a GWAS genome-wide significant variant in the GWAS Catalog. When the analysis was restricted to subclinical atherosclerosis traits, a locus was only considered novel for atherosclerosis if it was more than \u0026plusmn;\u0026thinsp;500 kb away from a GWAS genome-wide significant variant for CAD as well. A lead variant was considered to meet experiment-wide significance if it satisfied the Bonferroni-corrected threshold of P\u0026thinsp;\u0026lt;\u0026thinsp;3.3\u0026times;10\u003csup\u003e\u0026minus;\u0026thinsp;9\u003c/sup\u003e (5\u0026times;10\u003csup\u003e\u0026minus;\u0026thinsp;8\u003c/sup\u003e/15).\u003c/p\u003e \u003cp\u003eTo identify shared and unique loci across all multi-trait analyses, we merged loci into a single shared genomic locus if the lead variants from different analyses were within 500 kb of each other. For each shared locus, the variant with the most significant P-value across analyses was designated as the shared lead variant. Shared genomic locus IDs were retained to facilitate the identification of shared regions for colocalization analyses.\u003c/p\u003e\n\u003ch3\u003eTrait-trait and Trait- eQTL colocalization\u003c/h3\u003e\n\u003cp\u003eMulti-trait colocalization was conducted using HyPrColoc, a Bayesian divisive clustering algorithm designed to identify shared causal signals within a genomic region.\u003csup\u003e\u003cspan citationid=\"CR22\" class=\"CitationRef\"\u003e22\u003c/span\u003e\u003c/sup\u003e HyPrColoc also allows the clustering algorithm to be disabled, enabling colocalization analysis across pre-defined subsets of traits, similar to COLOC and MOLOC.\u003csup\u003e\u003cspan citationid=\"CR31\" class=\"CitationRef\"\u003e31\u003c/span\u003e\u003c/sup\u003e First, we performed trait-trait colocalization without clustering to replicate the multi-stage, multi-trait analysis conducted in N-GWAMA. Single-trait GWAS summary statistics were subset using a\u0026thinsp;\u0026plusmn;\u0026thinsp;500 kb window centered on the lead variant identified by N-GWAMA. Colocalization was then restricted to the traits included in the multi-trait GWAS analysis that identified the lead variant. Second, we conducted trait-trait colocalization across all traits using HyPrColoc\u0026rsquo;s clustering algorithm, focusing on the lead variant from the shared genomic locus. This analysis applied the variant-specific priors models with default parameters. Evidence for colocalization was determined based on the default variant- -specific regional and alignment priors (P\u003csub\u003e\u003cem\u003eR\u003c/em\u003e\u003c/sub\u003e\u0026lowast;=P\u003csub\u003e\u003cem\u003eA\u003c/em\u003e\u003c/sub\u003e\u003cem\u003e\u0026lowast;\u003c/em\u003e=0.5), with colocalization identified when P\u003csub\u003e\u003cem\u003eR\u003c/em\u003e\u003c/sub\u003eP\u003csub\u003e\u003cem\u003eA\u003c/em\u003e\u003c/sub\u003e\u0026ge;0.25. Strong evidence of colocalization across traits was defined as P\u003csub\u003eR\u003c/sub\u003eP\u003csub\u003eA\u003c/sub\u003e \u0026ge; 0.80.\u003c/p\u003e \u003cp\u003eFinally, for loci that colocalized in the trait-trait analysis, we performed trait-eQTL colocalization using eQTL data from GTEx v8 using the COLOC package in R.\u003csup\u003e\u003cspan citationid=\"CR31\" class=\"CitationRef\"\u003e31\u003c/span\u003e,32 33,\u003c/sup\u003e We leveraged summary statistics from the multi-trait GWAS for corresponding trait-trait colocalization and focused our analysis on a subset of GTEx v8 tissues selected based on their biological relevance to the studied traits (\u003cb\u003eSupplemental Table\u0026nbsp;8)\u003c/b\u003e. Colocalization was assessed using the default prior probabilities. Evidence for colocalization was defined as a posterior probability of a shared causal variant (PP\u003csub\u003eH4\u003c/sub\u003e)\u0026thinsp;\u0026ge;\u0026thinsp;0.50 and a conditional posterior probability (PP\u003csub\u003ec\u003c/sub\u003e=PP\u003csub\u003eH4\u003c/sub\u003e/(PP\u003csub\u003eH3\u003c/sub\u003e+PP\u003csub\u003eH4)\u003c/sub\u003e)\u0026thinsp;\u0026ge;\u0026thinsp;0.80. Strong evidence for colocalization was defined as (PP\u003csub\u003eH4\u003c/sub\u003e)\u0026thinsp;\u0026ge;\u0026thinsp;0.80 and a conditional posterior probability (PP\u003csub\u003eH4\u003c/sub\u003e/(PP\u003csub\u003eH3\u003c/sub\u003e+PP\u003csub\u003eH4\u003c/sub\u003e))\u0026thinsp;\u0026ge;\u0026thinsp;0.80.\u003c/p\u003e"},{"header":"Declarations","content":"\u003ch2\u003eConflict of interest\u003c/h2\u003e \u003cp\u003eDr. Clint L. Miller has received funding from AstraZeneca for unrelated work. All other authors have no conflicts of interest to declare.\u003c/p\u003e\u003ch2\u003eAcknowledgements\u003c/h2\u003e \u003cp\u003eThis study was funded by the National Institutes of Health (NIH) National Heart, Lung, and Blood Institute (NHLBI) grant number R01 HL146860 to P.S.dV. We thank GLGLC, ICBP and DIAMANTE consortia for making their summary statistics available to the research community. We also acknowledge funding support from NIH grants (R01HL148239, R01HL164577 and U01DK142283), Leducq Foundation network grant \u0026lsquo;COMET\u0026rsquo; (24CVD02), and American Heart Association Transformational Project Award (24TPA1300556) to C.L.M.\u003c/p\u003e"},{"header":"References","content":"\u003col\u003e\u003cli\u003e\u003cspan\u003eKlarin D et al (2017) Genetic analysis in UK Biobank links insulin resistance and transendothelial migration pathways to coronary artery disease. Nat Genet 49:1392\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eZhuang Z, Yao M, Wong JYY, Liu Z, Huang T (2021) Shared genetic etiology and causality between body fat percentage and cardiovascular diseases: a large-scale genome-wide cross-trait analysis. BMC Med 19:100\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eTada H et al (2014) Multiple associated variants increase the heritability explained for plasma lipids and coronary artery disease. Circ Cardiovasc Genet 7:583\u0026ndash;587\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eDeloukas P et al (2013) Large-scale association analysis identifies new risk loci for coronary artery disease. Nat Genet 45:25\u0026ndash;33\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eHowson JMM et al (2017) Fifteen new risk loci for coronary artery disease highlight arterial-wall-specific mechanisms. Nat Genet 49:1113\u0026ndash;1119\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eO'Donnell CJ et al (2011) Genome-wide association study for coronary artery calcification with follow-up in myocardial infarction. Circulation 124:2855\u0026ndash;2864\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eWebb TR et al (2017) Systematic Evaluation of Pleiotropy Identifies 6 Further Loci Associated With Coronary Artery Disease. J Am Coll Cardiol 69:823\u0026ndash;836\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eVinkhuyzen AAE, Wray NR, Yang J, Goddard ME, Visscher PM (2013) Estimation and Partition of Heritability in Human Populations Using Whole-Genome Analysis Methods. Annu Rev Genet 47:75\u0026ndash;95\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eLeBlanc M et al (2016) Identifying Novel Gene Variants in Coronary Artery Disease and Shared Genes With Several Cardiovascular Risk Factors. Circ Res 118:83\u0026ndash;94\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eAragam KG et al (2022) Discovery and systematic characterization of risk variants and genes for coronary artery disease in over a million participants. Nat Genet 54:1803\u0026ndash;1815\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eTcheandjieu C et al (2022) Large-scale genome-wide association study of coronary artery disease in genetically diverse populations. Nat Med 28:1679\u0026ndash;1692\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eEvangelou E et al (2018) Genetic analysis of over 1 million people identifies 535 new loci associated with blood pressure traits. Nat Genet 50:1412\u0026ndash;1425\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eWarren HR et al (2017) Genome-wide association analysis identifies novel blood pressure loci and offers biological insights into cardiovascular risk. Nat Genet 49:403\u0026ndash;415\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eGraham SE et al (2021) The power of genetic diversity in genome-wide association studies of lipids. Nature 600:675\u0026ndash;679\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eYeung MW et al (2022) Twenty-Five Novel Loci for Carotid Intima-Media Thickness: A Genome-Wide Association Study in \u0026gt;\u0026thinsp;45 000 Individuals and Meta-Analysis of \u0026gt;\u0026thinsp;100 000 Individuals. Arterioscler Thromb Vasc Biol 42:484\u0026ndash;501\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eMahajan A et al (2022) Multi-ancestry genetic study of type 2 diabetes highlights the power of diverse populations for discovery and translation. Nat Genet 54:560\u0026ndash;572\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eMorris AP et al (2012) Large-scale association analysis provides insights into the genetic architecture and pathophysiology of type 2 diabetes. Nat Genet 44:981\u0026ndash;990\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eScott RA et al (2017) An Expanded Genome-Wide Association Study of Type 2 Diabetes in Europeans. Diabetes 66:2888\u0026ndash;2902\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eKavousi M et al (2023) Multi-ancestry genome-wide study identifies effector genes and druggable pathways for coronary artery calcification. Nat Genet 55:1651\u0026ndash;1664\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eMeena D et al (2025) Genome-wide association study and multi-ancestry meta-analysis identify common variants associated with carotid artery intima-media thickness. \u003cem\u003emedRxiv\u003c/em\u003e, 04.11.25325582 (2025)\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eDai X, Wiernek S, Evans JP, Runge MS (2016) Genetics of coronary artery disease and myocardial infarction. World J Cardiol 8:1\u0026ndash;23\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eFoley CN et al (2021) A fast and efficient colocalization algorithm for identifying shared genetic risk factors across multiple traits. Nat Commun 12:764\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eBaselmans BML et al (2019) Multivariate genome-wide analyses of the well-being spectrum. Nat Genet 51:445\u0026ndash;451\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eZhao W et al (2017) Identification of new susceptibility loci for type 2 diabetes and shared etiological pathways with coronary heart disease. Nat Genet 49:1450\u0026ndash;1457\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eSiewert KM, Voight BF (2018) Bivariate Genome-Wide Association Scan Identifies 6 Novel Loci Associated With Lipid Levels and Coronary Artery Disease. Circ Genom Precis Med 11:e002239\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eLevin MG et al (2022) Genome-wide association and multi-trait analyses characterize the common genetic architecture of heart failure. Nat Commun 13:6914\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eBulik-Sullivan BK et al (2015) LD Score regression distinguishes confounding from polygenicity in genome-wide association studies. Nat Genet 47:291\u0026ndash;295\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eAuton A et al (2015) A global reference for human genetic variation. Nature 526:68\u0026ndash;74\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eWatanabe K, Taskesen E, van Bochoven A, Posthuma D (2017) Functional mapping and annotation of genetic associations with FUMA. Nat Commun 8:1826\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eCerezo M et al (2024) The NHGRI-EBI GWAS Catalog: standards for reusability, sustainability and diversity. Nucleic Acids Res 53:D998\u0026ndash;D1005\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eGiambartolomei C et al (2018) A Bayesian framework for multiple trait colocalization from summary association statistics. Bioinformatics 34:2538\u0026ndash;2545\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eBarbeira AN et al (2020) Exploiting the GTEx resources to decipher the mechanisms at GWAS loci. bioRxiv, 814350\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eThe Genotype-Tissue (2013) Expression (GTEx) project. Nat Genet 45:580\u0026ndash;585\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eFabregat A et al (2017) Reactome pathway analysis: a high-performance in-memory approach. BMC Bioinformatics 18:142\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eSenatus L et al (2023) DIAPH1 mediates progression of atherosclerosis and regulates hepatic lipid metabolism in mice. Commun Biol 6:280\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eRen Z et al (2020) Association of DIAPH1 gene polymorphisms with ischemic stroke. Aging 12:416\u0026ndash;435\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eRamasamy R, Shekhtman A, Schmidt AM (2023) RAGE/DIAPH1 and atherosclerosis through an evolving lens: Viewing the cell from the Inside - Out. Atherosclerosis, 117304\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eJones FS et al (2002) Regulation of vascular smooth muscle cell growth and adhesion by paired-related homeobox genes. Chest 121:89s\u0026ndash;90s\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eZhang X et al (2020) MicroRNA-217-5p ameliorates endothelial cell apoptosis induced by ox-LDL by targeting CLIC4. Nutr Metabolism Cardiovasc Dis 30:523\u0026ndash;533\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eHigashi Y (2022) Roles of Oxidative Stress and Inflammation in Vascular Endothelial Dysfunction-Related Disease. in \u003cem\u003eAntioxidants\u003c/em\u003e Vol. 11\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eKleinjan ML et al (2023) CLIC4 Regulates Endothelial Barrier Control by Mediating PAR1 Signaling via RhoA. Arterioscler Thromb Vasc Biol 43:1441\u0026ndash;1454\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eShao X, Liu Z, Liu S, Lin N, Deng Y (2021) Astragaloside IV alleviates atherosclerosis through targeting circ_0000231/miR-135a-5p/CLIC4 axis in AS cell model in vitro. Mol Cell Biochem 476:1783\u0026ndash;1795\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eTaliun D et al (2021) Sequencing of 53,831 diverse genomes from the NHLBI TOPMed Program. Nature 590:290\u0026ndash;299\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eChou W-C et al (2016) A combined reference panel from the 1000 Genomes and UK10K projects improved rare variant imputation in European and Chinese samples. Sci Rep 6:39313\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eMcCarthy S et al (2016) A reference panel of 64,976 haplotypes for genotype imputation. Nat Genet 48:1279\u0026ndash;1283\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eBulik-Sullivan B et al (2015) An atlas of genetic correlations across human diseases and traits. Nat Genet 47:1236\u0026ndash;1241\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eWang K, Li M, Hakonarson H (2010) ANNOVAR: functional annotation of genetic variants from high-throughput sequencing data. Nucleic Acids Res 38:e164\u0026ndash;e164\u003c/span\u003e\u003c/li\u003e\u003c/ol\u003e"}],"fulltextSource":"","fullText":"","funders":[],"hasAdminPriorityOnWorkflow":false,"hasManuscriptDocX":true,"hasOptedInToPreprint":true,"hasPassedJournalQc":"","hasAnyPriority":true,"hideJournal":true,"highlight":"","institution":"","isAcceptedByJournal":false,"isAuthorSuppliedPdf":false,"isDeskRejected":"","isHiddenFromSearch":false,"isInQc":false,"isInWorkflow":false,"isPdf":false,"isPdfUpToDate":true,"isWithdrawnOrRetracted":false,"journal":{"display":true,"email":"[email protected]","identity":"researchsquare","isNatureJournal":false,"hasQc":true,"allowDirectSubmit":true,"externalIdentity":"","sideBox":"","snPcode":"","submissionUrl":"/submission","title":"Research Square","twitterHandle":"researchsquare","acdcEnabled":true,"dfaEnabled":false,"editorialSystem":"","reportingPortfolio":"","inReviewEnabled":false,"inReviewRevisionsEnabled":true},"keywords":"","lastPublishedDoi":"10.21203/rs.3.rs-6456056/v1","lastPublishedDoiUrl":"https://doi.org/10.21203/rs.3.rs-6456056/v1","license":{"name":"CC BY 4.0","url":"https://creativecommons.org/licenses/by/4.0/"},"manuscriptAbstract":"\u003cp\u003eMeasures of subclinical atherosclerosis, such as coronary artery calcification (CAC) and carotid intima-media thickness (CIMT), reflect the underlying pathophysiology of coronary artery disease (CAD) and are genetically correlated with CAD and related risk factors. \u0026nbsp;Leveraging summary statistics from genome-wide association studies of CAD, CIMT, CAC, type 2 diabetes, low-density lipoprotein cholesterol, and systolic blood pressure, we performed 15 separate multi-trait GWAS to identify shared susceptibility loci and elucidate the pleiotropic architecture underlying atherosclerosis. We identified 442 shared risk loci across all analyses that met an experiment-wide Bonferroni threshold of 3.3 × 10\u003csup\u003e-9\u003c/sup\u003e, uncovering 195 novel atherosclerosis loci. Multi-trait colocalization confirmed a shared causal signal in 25 shared novel loci for atherosclerosis. Trait-eQTL colocalization identified evidence of a shared causal signal in arterial, subcutaneous adipose, and cardiac tissues, implicating genes such as \u003cem\u003ePRRX2\u003c/em\u003e, \u003cem\u003eBNC2\u003c/em\u003e, \u003cem\u003eCLIC4\u003c/em\u003e, \u003cem\u003eSCAI\u003c/em\u003e, and \u003cem\u003ePPP6C\u003c/em\u003e, and pathways related to vascular remodeling, inflammation, and metabolic regulation.\u0026nbsp;\u003c/p\u003e","manuscriptTitle":"A multi-trait genome-wide association study of coronary artery disease and subclinical atherosclerosis traits","msid":"","msnumber":"","nonDraftVersions":[{"code":1,"date":"2025-04-21 05:03:31","doi":"10.21203/rs.3.rs-6456056/v1","editorialEvents":[{"type":"communityComments","content":0}],"status":"published","journal":{"display":true,"email":"[email protected]","identity":"researchsquare","isNatureJournal":false,"hasQc":true,"allowDirectSubmit":true,"externalIdentity":"","sideBox":"","snPcode":"","submissionUrl":"/submission","title":"Research Square","twitterHandle":"researchsquare","acdcEnabled":true,"dfaEnabled":false,"editorialSystem":"","reportingPortfolio":"","inReviewEnabled":false,"inReviewRevisionsEnabled":true}}],"origin":"","ownerIdentity":"10b80eb5-ee71-4191-8e82-056adb1ff40a","owner":[],"postedDate":"April 21st, 2025","published":true,"recentEditorialEvents":[],"rejectedJournal":[],"revision":"","amendment":"","status":"posted","subjectAreas":[{"id":47364238,"name":"Health sciences/Cardiology/Cardiovascular biology/Cardiovascular genetics"},{"id":47364239,"name":"Health sciences/Cardiology/Cardiovascular biology/Cardiovascular diseases/Vascular diseases/Atherosclerosis"}],"tags":[],"updatedAt":"2026-01-14T05:53:33+00:00","versionOfRecord":[],"versionCreatedAt":"2025-04-21 05:03:31","video":"","vorDoi":"","vorDoiUrl":"","workflowStages":[]},"version":"v1","identity":"rs-6456056","journalConfig":"researchsquare"},"__N_SSP":true},"page":"/article/[identity]/[[...version]]","query":{"redirect":"/article/rs-6456056","identity":"rs-6456056","version":["v1"]},"buildId":"8U1c8b4HqxoKbykW_rLl7","isFallback":false,"isExperimentalCompile":false,"dynamicIds":[84888],"gssp":true,"scriptLoader":[]}