Search

Background Genetic association studies aim to test for disease or trait association with genetic variants, either throughout the human genome or in regions of interest. However, for most diseases and traits, the combined effects of associated genetic variants explain only a small proportion of the genetic variation. This “missing heritability” may be a result of the small effects of common variants considered in the genetic association studies. Rare variants may also play an important role in understanding the missing heritability of complex traits. Method We propose a novel weight-adjustment approach to combine gene expression into rare variant analysis. Results from previous simulation studies suggested that incorporating gene expression information can lead to substantial gain in statistical power. Results Using the family data set provided through the Genetic Analysis Workshop 19, we identified susceptible genes associated with blood pressure regulation. Conclusions These findings provide valuable information for further functional studies for blood pressure control and mechanism.

Pneumatization of the maxillary sinus limits the quantity of alveolar bone available for implant placement and may result in a lack of primary stability and difficulty in achieving osseointegration. The purpose of this study was to retrospectively analyze a group of patients who had implants placed in the posterior maxilla, calculate the prevalence of sinus augmentation, and identify factors related to sinus augmentation. With institutional review board approval, dental records from a population of patients who had implants placed in the maxillary posterior region between January 2000 and December 2004 were used to create a database. Independent variables were classified as continuous (age of the patient at stage 1 implant surgery [S1], time between extraction and S1, time between extraction and sinus augmentation, and time between sinus augmentation and S1) and categorical (gender, implant failure, American Society of Anesthesiologists system classification, smoking, osteoporosis, residual crestal bone height, implant position, implant proximity, prostheses type, and implant diameter and length). The dependent variable was the incidence of a sinus augmentation procedure. Simple logistic regression was used to assess the influence of each factor on the presence of sinus augmentation (P < .05). The final database included 502 maxillary posterior implants with an overall survival rate of 93.2% over a mean follow-up period of 35.7 months. Of 502 implants, 272 (54.2%) were associated with a sinus augmentation procedure. Among variables, residual crestal bone height (P < .001), implant position (P < .001), implant proximity (P < .001), prosthesis type (P < .001), implant failure (P < .01), and implant diameter (P < .01), were statistically associated with sinus augmentation. Within the limitations of this retrospective study, the results suggest that more than half (54.2%) of the maxillary posterior implants were involved with a sinus augmentation procedure. The prevalence of sinus augmentation increased with decreased residual crestal bone height, more posterior implant locations, and complete or partial edentulism. Sinus augmentation was significantly associated with implant failure and wide implants.

Venous thromboembolism (VTE) is a common, heritable disease resulting in high rates of hospitalization and mortality. Yet few associations between VTE and genetic variants, all in the coagulation pathway, have been established. To identify additional genetic determinants of VTE, we conducted a two-stage genome-wide association study (GWAS) among individuals of European ancestry in the extended cohorts for heart and aging research in genomic epidemiology (CHARGE) VTE consortium. The discovery GWAS comprised 1,618 incident VTE cases out of 44,499 participants from six community-based studies. Genotypes for genome-wide single-nucleotide polymorphisms (SNPs) were imputed to approximately 2.5 million SNPs in HapMap and association with VTE assessed using study-design appropriate regression methods. Meta-analysis of these results identified two known loci, in F5 and ABO. Top 1,047 tag SNPs (P ≤ 0.0016) from the discovery GWAS were tested for association in an additional 3,231 cases and 3,536 controls from three case-control studies. In the combined data from these two stages, additional genome-wide significant associations were observed on 4q35 at F11 (top SNP rs4253399, intronic to F11) and on 4q28 at FGG (rs6536024, 9.7 kb from FGG; P < 5.0 × 10(-13) for both). The associations at the FGG locus were not completely explained by previously reported variants. Loci at or near SUSD1 and OTUD7A showed borderline yet novel associations (P < 5.0 × 10(-6) ) and constitute new candidate genes. In conclusion, this large GWAS replicated key genetic associations in F5 and ABO, and confirmed the importance of F11 and FGG loci for VTE. Future studies are warranted to better characterize the associations with F11 and FGG and to replicate the new candidate associations.

Identifying gene-environment (G-E) interactions can contribute to a better understanding of disease etiology, which may help researchers develop disease prevention strategies and interventions. One big criticism of studying G-E interaction is the lack of power due to sample size. Studies often restrict the interaction search to the top few hundred hits from a genome-wide association study or focus on potential candidate genes. In this paper, we test interactions between a candidate gene and an environmental factor to improve power by analyzing multiple variants within a gene. We extend recently developed score statistic based genetic association testing approaches to the G-E interaction testing problem. We also propose tests for interaction using gene-based summary measures that pool variants together. Although it has recently been shown that these summary measures can be biased and may lead to inflated type I error, we show that under several realistic scenarios, we can still provide valid tests of interaction. These tests use significantly less degrees of freedom and thus can have much higher power to detect interaction. Additionally, we demonstrate that the iSeq-aSum-min test, which combines a gene-based summary measure test, iSeq-aSum-G, and an interaction-based summary measure test, iSeq-aSum-I, provides a powerful alternative to test G-E interaction. We demonstrate the performance of these approaches using simulation studies and illustrate their performance to study interaction between the SNPs in several candidate genes and family climate environment on alcohol consumption using the Minnesota Center for Twin and Family Research dataset.

Background— Fibrinogen is both central to blood coagulation and an acute-phase reactant. We aimed to identify common variants influencing circulation fibrinogen levels. Methods and Results— We conducted a genome-wide association analysis on 6 population-based studies, the Rotterdam Study, the Framingham Heart Study, the Cardiovascular Health Study, the Atherosclerosis Risk in Communities Study, the Monitoring of Trends and Determinants in Cardiovascular Disease/KORA Augsburg Study, and the British 1958 Birth Cohort Study, including 22 096 participants of European ancestry. Four loci were marked by 1 or more single-nucleotide polymorphisms that demonstrated genome-wide significance ( P −8 ). These included a single-nucleotide polymorphism located in the fibrinogen β chain ( FGB ) gene and 3 single-nucleotide polymorphisms representing newly identified loci. The high-signal single-nucleotide polymorphisms were rs1800789 in exon 7 of FGB ( P =1.8�10 −30 ), rs2522056 downstream from the interferon regulatory factor 1 ( IRF1 ) gene ( P =1.3�10 −15 ), rs511154 within intron 1 of the propionyl coenzyme A carboxylase ( PCCB ) gene ( P =5.9�10 −10 ), and rs1539019 on the NLR family pyrin domain containing 3 isoforms ( NLRP3 ) gene ( P =1.04�10 −8 ). Conclusions— Our findings highlight biological pathways that may be important in regulation of inflammation underlying cardiovascular disease.

Low von Willebrand factor (VWF) levels are associated with bleeding symptoms and are a diagnostic criterion for von Willebrand disease, the most common inherited bleeding disorder. To date, it is unclear which genetic loci are associated with reduced VWF levels. Therefore, we conducted a meta-analysis of genome-wide association studies to identify genetic loci associated with low VWF levels. For this meta-analysis, we included 31 149 participants of European ancestry from 11 community-based studies. From all participants, VWF antigen (VWF:Ag) measurements and genome-wide single-nucleotide polymorphism (SNP) scans were available. Each study conducted analyses using logistic regression of SNPs on dichotomized VWF: Ag measures (lowest 5% for blood group O and non-O) with an additive genetic model adjusted for age and sex. An inverse-variance weighted meta-analysis was performed for VWF:Ag levels. A total of 97 SNPs exceeded the genome-wide significance threshold of 5 x 10(-8) and comprised five loci on four different chromosomes: 6q24 (smallest P-value 5.8 x 10(-10)), 9q34 (2.4 x 10(-64)), 12p13 (5.3x 10(-22)), 12q23 (1.2 x 10(-8)) and 13q13 (2.6 x 10(-8)). All loci were within or close to genes, including STXBP5 (Syntaxin Binding Protein 5) (6q24), STAB5 (stabilin-5) (12q23), ABO (9q34), VWF (12p13) and UFM1 (ubiquitin-fold modifier 1) (13q13). Of these, UFM1 has not been previously associated with VWF: Ag levels. Four genes that were previously associated with VWF levels (VWF, ABO, STXBP5 and STAB2) were also associated with low VWF levels, and, in addition, we identified a new gene, UFM1, that is associated with low VWF levels. These findings point to novel mechanisms for the occurrence of low VWF levels.

Genome-wide association studies have previously identified 23 genetic loci associated with circulating fibrinogen concentration. These studies used HapMap imputation and did not examine the X-chromosome. 1000 Genomes imputation provides better coverage of uncommon variants, and includes indels. We conducted a genome-wide association analysis of 34 studies imputed to the 1000 Genomes Project reference panel and including similar to 120 000 participants of European ancestry (95 806 participants with data on the X-chromosome). Approximately 10.7 million single-nucleotide polymorphisms and 1.2 million indels were examined. We identified 41 genome-wide significant fibrinogen loci ; of which, 18 were newly identified. There were no genome-wide significant signals on the X-chromosome. The lead variants of five significant loci were indels. We further identified six additional independent signals, including three rare variants, at two previously characterized loci: FGB and IRF1. Together the 41 loci explain 3% of the variance in plasma fibrinogen concentration.

Activated partial thromboplastin time (aPTT) and prothrombin time (PT) are clinical tests commonly used to screen for coagulation-factor deficiencies. One genome-wide association study (GWAS) has been reported previously for aPTT, but no GWAS has been reported for PT. We conducted a GWAS and meta-analysis to identify genetic loci for aPTT and PT. The GWAS for aPTT was conducted in 9,240 individuals of European ancestry from the Atherosclerosis Risk in Communities (ARIC) study, and the GWAS for PT was conducted in 2,583 participants from the Genetic Study of Three Population Microisolates in South Tyrol (MICROS) and the Lothian Birth Cohorts (LBC) of 1921 and 1936. Replication was assessed in 1,041 to 3,467 individuals. For aPTT, previously reported associations with KNG1, HRG, F11, F12, and ABO were confirmed. A second independent association in ABO was identified and replicated (rs8176704, p = 4.26 × 10(-24)). Pooling the ARIC and replication data yielded two additional loci in F5 (rs6028, p = 3.22 × 10(-9)) and AGBL1 (rs2469184, p = 3.61 × 10(-8)). For PT, significant associations were identified and confirmed in F7 (rs561241, p = 3.71 × 10(-56)) and PROCR/EDEM2 (rs2295888, p = 5.25 × 10(-13)). Assessment of existing gene expression and coronary artery disease (CAD) databases identified associations of five of the GWAS loci with altered gene expression and two with CAD. In summary, eight genetic loci that account for ∼29% of the variance in aPTT and two loci that account for ∼14% of the variance in PT were detected and supported by functional data.

We conducted a genome-wide association study to identify novel associations between genetic variants and circulating plasminogen activator inhibitor-1 (PAI-1) concentration, and examined functional implications of variants and genes that were discovered. A discovery meta-analysis was performed in 19 599 subjects, followed by replication analysis of genome-wide significant (P < 5 × 10−8) single nucleotide polymorphisms (SNPs) in 10 796 independent samples. We further examined associations with type 2 diabetes and coronary artery disease, assessed the functional significance of the SNPs for gene expression in human tissues, and conducted RNA-silencing experiments for one novel association. We confirmed the association of the 4G/5G proxy SNP rs2227631 in the promoter region of SERPINE1 (7q22.1) and discovered genome-wide significant associations at 3 additional loci: chromosome 7q22.1 close to SERPINE1 (rs6976053, discovery P = 3.4 × 10−10); chromosome 11p15.2 within ARNTL (rs6486122, discovery P = 3.0 × 10−8); and chromosome 3p25.2 within PPARG (rs11128603, discovery P = 2.9 × 10−8). Replication was achieved for the 7q22.1 and 11p15.2 loci. There was nominal association with type 2 diabetes and coronary artery disease at ARNTL (P < .05). Functional studies identified MUC3 as a candidate gene for the second association signal on 7q22.1. In summary, SNPs in SERPINE1 and ARNTL and an SNP associated with the expression of MUC3 were robustly associated with circulating levels of PAI-1.

In multi-locus association analysis, since some markers may not be associated with a trait, it seems attractive to use penalized regression with the capability of automatic variable selection. On the other hand, in spite of a rapidly growing body of literature on penalized regression, most focus on variable selection and outcome prediction, for which penalized methods are generally more effective than their non-penalized counterparts. However, for statistical inference, i.e. hypothesis testing and interval estimation, it is less clear how penalized methods would perform, or even how to best apply them, largely due to lack of studies on this topic. In our motivating data for a cohort of kidney transplant recipients, it is of primary interest to assess whether a group of genetic variants are associated with a binary clinical outcome, acute rejection at 6 months. In this paper, we study some technical issues and alternative implementations of hypothesis testing in Lasso penalized logistic regression, and compare their performance with each other and with several existing global tests, some of which are specifically designed as variance component tests for high-dimensional data. The most interesting, and perhaps surprising, conclusion of this study is that, for low to moderately high-dimensional data, statistical tests based on Lasso penalized regression are not necessarily more powerful than some existing global tests. In addition, in penalized regression, rather than building a test based on a single selected “best” model, combining multiple tests, each of which is built on a candidate model, might be more promising.

In anticipation of the availability of next-generation sequencing data, there is increasing interest in investigating association between complex traits and rare variants (RVs). In contrast to association studies for common variants (CVs), due to the low frequencies of RVs, common wisdom suggests that existing statistical tests for CVs might not work, motivating the recent development of several new tests for analyzing RVs, most of which are based on the idea of pooling/collapsing RVs. However, there is a lack of evaluations of, and thus guidance on the use of, existing tests. Here we provide a comprehensive comparison of various statistical tests using simulated data. We consider both independent and correlated rare mutations, and representative tests for both CVs and RVs. As expected, if there are no or few non-causal (i.e. neutral or non-associated) RVs in a locus of interest while the effects of causal RVs on the trait are all (or mostly) in the same direction (i.e. either protective or deleterious, but not both), then the simple pooled association tests (without selecting RVs and their association directions) and a new test called kernel-based adaptive clustering (KBAC) perform similarly and are most powerful; KBAC is more robust than simple pooled association tests in the presence of non-causal RVs; however, as the number of non-causal CVs increases and/or in the presence of opposite association directions, the winners are two methods originally proposed for CVs and a new test called C-alpha test proposed for RVs, each of which can be regarded as testing on a variance component in a random-effects model. Interestingly, several methods based on sequential model selection (i.e. selecting causal RVs and their association directions), including two new methods proposed here, perform robustly and often have statistical power between those of the above two classes.

Studying one locus or one single nucleotide polymorphism (SNP) at a time may not be sufficient to understand complex diseases because they are unlikely to result from the effect of only one SNP. Each SNP alone may have little or no effect on the risk of the disease, but together they may increase the risk substantially. Analyses focusing on individual SNPs ignore the possibility of interaction among SNPs. In this paper, we propose a parsimonious model to assess the joint effect of a group of SNPs in a case-control study. The model implements a data reduction strategy within a likelihood framework and uses a test to assess the statistical significance of the effect of the group of SNPs on the binary trait. The primary advantage of the proposed approach is that the dimension reduction technique produces a test statistic with degrees of freedom significantly lower than a multiple logistic regression with only main effects of the SNPs, and our parsimonious model can incorporate the possibility of interaction among the SNPs. Moreover, the proposed approach estimates the direction of association of each SNP with the disease and provides an estimate of the average effect of the group of SNPs positively and negatively associated with the disease in the given SNP set. We illustrate the proposed model on simulated and real data, and compare its performance with a few other existing approaches. Our proposed approach appeared to outperform the other approaches for independent SNPs in our simulation studies.

Genome-wide association studies (GWAS) using family data involve association analyses between hundreds of thousands of markers and a trait for a large number of related individuals. The correlations among relatives bring statistical and computational challenges when performing these large-scale association analyses. Recently, several rapid methods accounting for both within- and between-family variation have been proposed. However, these techniques mostly model the phenotypic similarities in terms of genetic relatedness. The familial resemblances in many family-based studies such as twin studies are not only due to the genetic relatedness, but also derive from shared environmental effects and assortative mating. In this paper, we propose 2 generalized least squares (GLS) models for rapid association analysis of family-based GWAS, which accommodate both genetic and environmental contributions to familial resemblance. In our first model, we estimated the joint genetic and environmental variations. In our second model, we estimated the genetic and environmental components separately. Through simulation studies, we demonstrated that our proposed approaches are more powerful and computationally efficient than a number of existing methods are. We show that estimating the residual variance-covariance matrix in the GLS models without SNP effects does not lead to an appreciable bias in the p values as long as the SNP effect is small (i.e. accounting for no more than 1% of trait variance).

Protein C is an important endogenous anticoagulant in hemostasis. Deficiencies of protein C due to genetic mutations or a low level of circulating protein C increase the risk of venous thromboembolism. We performed a genome-wide association scan for plasma protein C antigen concentration with approximately 2.5 million single-nucleotide polymorphisms in 8048 individuals of European ancestry and a replication analysis in a separate sample of 1376 individuals in the Atherosclerosis Risk in Communities Study. Four independent loci from 3 regions were identified with genome-wide significance: 2p23 (GCKR, best SNP rs1260326, P = 2.04 × 10−17), 2q13-q14 (PROC, rs1158867, P = 3.77 × 10−36), 20q11 (near and within PROCR, rs8119351, P = 2.68 × 10−203), and 20q11.22 (EDEM2, rs6120849, P = 7.19 × 10−37 and 5.23 × 10−17 before and after conditional analysis, respectively). All 4 loci replicated in the independent sample. Furthermore, pooling the discovery and replication sets yielded an additional locus at chromosome 7q11.23 (BAZ1B, rs17145713, P = 2.83 × 10−8). The regions marked by GCKR, EDEM2, and BAZ1B are novel loci that have not been previously reported for association with protein C concentration. In summary, this first genome-wide scan for circulating protein C concentration identified both new and known loci in the general population. These findings may improve the understanding of physiologic mechanisms in protein C regulation.

Graft-versus-host disease (GVHD) is a frequent complication after hematopoietic cell transplant (HCT). Tissue damage as a result of chemoradiation injury is the initiating event in the pathogenesis of acute GVHD. Variations in DNA repair can influence the amount of tissue damage in response to alkylating agents and ionizing radiation used as conditioning during HCT. As DNA damage caused by these agents is repaired by the base excision repair (BER) pathway, we hypothesized that single-nucleotide polymorphisms (SNPs) in BER pathway will be associated with GVHD after HCT. Hence, we analyzed 179 SNPs in BER pathway in 470 recipients of allogeneic HCT for association with acute and chronic GVHD. In multivariate analysis, one SNP (rs6844176) in RFC1 (replication factor C (activator 1)) gene was independently associated with a higher risk of grade II-IV acute GVHD (relative risk (RR): 1.39, 95% confidence interval (CI): 1.14-1.70, P=0.001), and showed a trend toward higher risk of grade III-IV acute GVHD (RR: 1.33, 95% CI: 0.95-1.85, P=0.09). One SNP in PARP1 gene (rs1805410) was associated with a higher risk of chronic GVHD (RR: 1.81, 95% CI: 1.29-2.54, P=0.001). These results show that SNPs in the BER pathway can be used as genetic biomarkers to predict those at high risk for GVHD toward whom novel prophylactic strategies could be targeted.

Osteosarcoma is considered to be the most common primary malignant bone cancer among children and young adults. Previous studies suggest growth spurts and height to be risk factors for osteosarcoma. However, studies on the genetic cause are still limited given the rare occurrence of the disease. In this study, we investigated in a family trio data set that is composed of 209 patients and their unaffected parents and conducted a genome-wide association study (GWAS) to identify genetic risk factors for osteosarcoma. We performed a Bayesian gene-based GWAS based on the single-nucleotide polymorphism (SNP)-level summary statistics obtained from a likelihood ratio test of the trio data, which uses a hierarchically structured prior that incorporates the SNP-gene hierarchical structure. The Bayesian approach has higher power than SNP-level GWAS analysis due to the reduced number of tests and is robust by accounting for the correlations between SNPs so that it borrows information across SNPs within a gene. We identified 217 genes that achieved genome-wide significance. Ingenuity pathway analysis of the gene set indicated that osteosarcoma is potentially related to TP53, estrogen receptor signaling, xenobiotic metabolism signaling, and RANK signaling in osteoclasts.

Mycophenolic acid (MPA) is glucuronidated primarily by uridine diphosphate glucuronosyltransferase enzymes (UGT) 1A9 and 1A8. These enzymes are highly polymorphic resulting in low activity and high expression phenotypes. We hypothesized that polymorphisms of UGT1A9 and 1A8 may alter MPA pharmacokinetics in kidney transplantation.One hundred seventeen kidney (n = 93), pancreas (n = 11), or simultaneous kidney and pancreas (SPK) (n = 13) transplant recipients were studied for the effect of UGT1A9 and UGT1A8 polymorphisms on MPA dose-corrected trough concentrations. Individuals were genotyped for UGT1A8 and UGT1A9 polymorphisms (1A8*2, 1A8*3, 1A9*3, 1A9-275 and 1A9-2152). Linear regression was used to estimate the effect of UGT polymorphisms on the individual's mean MPA dose-corrected trough concentration with and without stratification by calcineurin inhibitor. A multiple linear regression analysis was performed to assess the dependence between the average MPA dose-corrected trough concentration and age, gender, UGT genotype (1A8*2, 1A8*3, 1A9*3, 1A9-275, 1A9-2152), serum albumin, hemoglobin (Hgb), hematocrit (HCT), liver transaminases (AST, ALT), serum creatinine, and bilirubin.Mycophenolic acid dose-corrected trough concentrations were 60% higher in subjects heterozygous or homozygous for UGT1A8*2 than in those with the wild type (p = 0.02); however, this effect was dependent on concomitant calcineurin inhibitor. When subjects were stratified by calcineurin inhibitor status, the UGT1A8*2 effect was only apparent in the tacrolimus group (p0.01). Mycophenolic acid dose-corrected trough concentrations were 70% lower in carriers of the UGT1A9 -275TA/-2152 CT polymorphism who received cyclosporine (p0.01). There was no effect of the UGT1A9 -275TA/-2152CT polymorphism in the tacrolimus group.The effect of UGT1A8 and UGT1A9 variants on MPA metabolism appears to be modified by concomitant calcineurin inhibitor therapy. Confirmatory in vivo and in vitro studies are needed.

Although serum total immunoglobulin E (IgE) is generally elevated in atopic conditions, it is an unreliable trait for dissecting the genetic and environmental components contributing to atopic immune responses, because it can be significantly confounded by demographic factors (age, gender, and race) and clinical status (atopic vs nonatopic). Allergen-specific IgE is a discontinuous trait present only in those with sensitivity to allergens. However, all people will produce allergen-specific immunoglobulin G1 (IgG1), which is elevated among those atopically sensitized to specific allergens. We screened 91 Caucasian nuclear families (N = 367) with medical histories of atopic diseases and used variance components analysis to compare heritability estimates for total IgE and IgG1 produced against the common major allergen from house dust mite Dermatophagoides pteronyssinus (Der p 1). An estimate of total IgE heritability was about 48%, although this was significantly confounded by age, gender, and clinical atopic status. In contrast, Der p 1–IgG1 demonstrated a significant inherited component of about 62% that was not influenced by age, gender, or clinical status. For genetic studies of atopic humoral responses, allergen-specific IgG1 may be a more reliable quantitative trait than serum IgE. Moreover, atopy is an inherited deregulation of immune responses to noninfectious antigens, involving antibody isotypes other than IgE.

International audience; Venous thromboembolism (VTE), the third leading cause of cardiovascular mortality, is a complex thrombotic disorder with environmental and genetic determinants. Although several genetic variants have been found associated with VTE, they explain a minor proportion of VTE risk in cases. We undertook a meta-analysis of genome-wide association studies (GWASs) to identify additional VTE susceptibility genes. Twelve GWASs totaling 7,507 VTE case subjects and 52,632 control subjects formed our discovery stage where 6,751,884 SNPs were tested for association with VTE. Nine loci reached the genome-wide significance level of 5 × 10(-8) including six already known to associate with VTE (ABO, F2, F5, F11, FGG, and PROCR) and three unsuspected loci. SNPs mapping to these latter were selected for replication in three independent case-control studies totaling 3,009 VTE-affected individuals and 2,586 control subjects. This strategy led to the identification and replication of two VTE-associated loci, TSPAN15 and SLC44A2, with lead risk alleles associated with odds ratio for disease of 1.31 (p = 1.67 × 10(-16)) and 1.21 (p = 2.75 × 10(-15)), respectively. The lead SNP at the TSPAN15 locus is the intronic rs78707713 and the lead SLC44A2 SNP is the non-synonymous rs2288904 previously shown to associate with transfusion-related acute lung injury. We further showed that these two variants did not associate with known hemostatic plasma markers. TSPAN15 and SLC44A2 do not belong to conventional pathways for thrombosis and have not been associated to other cardiovascular diseases nor related quantitative biomarkers. Our findings uncovered unexpected actors of VTE etiology and pave the way for novel mechanistic concepts of VTE pathophysiology.

Genome-wide Association Studies (GWASs) for complex diseases often collect data on multiple correlated endo-phenotypes. Multivariate analysis of these correlated phenotypes can improve the power to detect genetic variants. Multivariate analysis of variance (MANOVA) can perform such association analysis at a GWAS level, but the behavior of MANOVA under different trait models has not been carefully investigated. In this paper, we show that MANOVA is generally very powerful for detecting association but there are situations, such as when a genetic variant is associated with all the traits, where MANOVA may not have any detection power. In these situations, marginal model based methods, however, perform much better than multivariate methods. We investigate the behavior of MANOVA, both theoretically and using simulations, and derive the conditions where MANOVA loses power. Based on our findings, we propose a unified score-based test statistic USAT that can perform better than MANOVA in such situations and nearly as well as MANOVA elsewhere. Our proposed test reports an approximate asymptotic p-value for association and is computationally very efficient to implement at a GWAS level. We have studied through extensive simulations the performance of USAT, MANOVA and other existing approaches and demonstrated the advantage of using the USAT approach to detect association between a genetic variant and multivariate phenotypes. We applied USAT to data from three correlated traits collected on 5, 816 Caucasian individuals from the Atherosclerosis Risk in Communities (ARIC, The ARIC Investigators [1989]) Study and detected some interesting associations.

Coagulation factor VIII and von Willebrand factor (VWF) are key proteins in procoagulant activation. Higher FVIII coagulant activity (FVIII :C) and VWF antigen (VWF :Ag) are risk factors for cardiovascular disease and venous thromboembolism. Beyond associations with ABO blood group, genetic determinants of FVIII and VWF are not well understood, especially in non European-American populations. We performed a genetic association study of FVIII :C and VWF:Ag that assessed 50,000 gene-centric single nucleotide polymorphisms (SNPs) in 18,556 European Americans (EAs) and 5,047 African Americans (AAs) from five population-based cohorts. Previously unreported associations for FVIII :C were identified in both AAs and EAs with KNG1 (most significantly associated SNP rs710446, Ile581Thr, Ile581Thr, P = 5.10 × 10(-7) in EAs and P = 3.88 × 10(-3) in AAs) and VWF rs7962217 (Gly2705Arg,P = 6.30 × 10(-9) in EAs and P = 2.98 × 10(-2) in AAs. Significant associations for FVIII :C were also observed with F8/TMLHE region SNP rs12557310 in EAs (P = 8.02 × 10(-10) ), with VWF rs1800380 in AAs (P = 5.62 × 10(-11) ), and with MAT1A rs2236568 in AAs (P51.69 × 10(-6) ). We replicated previously reported associations of FVIII :C and VWF :Ag with the ABO blood group, VWF rs1063856(Thr789Ala), rs216321 (Ala852Gln), and VWF rs2229446 (Arg2185Gln). Findings from this study expand our understanding of genetic influences for FVIII :C and VWF :Ag in both EAs and AAs.

We carried out a genome-wide association study (GWAS) for general cognitive ability (GCA) plus three other analyses of GWAS data that aggregate the effects of multiple single-nucleotide polymorphisms (SNPs) in various ways. Our multigenerational sample comprised 7,100 Caucasian participants, drawn from two longitudinal family studies, who had been assessed with an age-appropriate IQ test and had provided DNA samples passing quality screens. We conducted the GWAS across ∼ 2.5 million SNPs (both typed and imputed), using a generalized least-squares method appropriate for the different family structures present in our sample, and subsequently conducted gene-based association tests. We also conducted polygenic prediction analyses under five-fold cross-validation, using two different schemes of weighting SNPs. Using parametric bootstrapping, we assessed the performance of this prediction procedure under the null. Finally, we estimated the proportion of variance attributable to all genotyped SNPs as random effects with software GCTA. The study is limited chiefly by its power to detect realistic single-SNP or single-gene effects, none of which reached genome-wide significance, though some genomic inflation was evident from the GWAS. Unit SNP weights performed about as well as least-squares regression weights under cross-validation, but the performance of both increased as more SNPs were included in calculating the polygenic score. Estimates from GCTA were 35% of phenotypic variance at the recommended biological-relatedness ceiling. Taken together, our results concur with other recent studies: they support a substantial heritability of GCA, arising from a very large number of causal SNPs, each of very small effect. We place our study in the context of the literature-both contemporary and historical-and provide accessible explication of our statistical methods.

Multi-locus effect modeling is a powerful approach for detection of genes influencing a complex disease. Especially for rare variants, we need to analyze multiple variants together to achieve adequate power for detection. In this paper, we propose several parsimonious branching model techniques to assess the joint effect of a group of rare variants in a case-control study. These models implement a data reduction strategy within a likelihood framework and use a weighted score test to assess the statistical significance of the effect of the group of variants on the disease. The primary advantage of the proposed approach is that it performs model-averaging over a substantially smaller set of models supported by the data and thus gains power to detect multi-locus effects. We illustrate these proposed approaches on simulated and real data and study their performance compared to several existing rare variant detection approaches. The primary goal of this paper is to assess if there is any gain in power to detect association by averaging over a number of models instead of selecting the best model. Extensive simulations and real data application demonstrate the advantage the proposed approach in presence of causal variants with opposite directional effects along with a moderate number of null variants in linkage disequilibrium.

Coagulation factor VIII and von Willebrand factor (VWF) are key proteins in procoagulant activation. Higher FVIII coagulant activity (FVIII :C) and VWF antigen (VWF :Ag) are risk factors for cardiovascular disease and venous thromboembolism. Beyond associations with ABO blood group, genetic determinants of FVIII and VWF are not well understood, especially in non European-American populations. We performed a genetic association study of FVIII :C and VWF:Ag that assessed 50,000 gene-centric single nucleotide polymorphisms (SNPs) in 18,556 European Americans (EAs) and 5,047 African Americans (AAs) from five population-based cohorts. Previously unreported associations for FVIII :C were identified in both AAs and EAs with KNG1 (most significantly associated SNP rs710446, Ile581Thr, Ile581Thr, P = 5.10 × 10(-7) in EAs and P = 3.88 × 10(-3) in AAs) and VWF rs7962217 (Gly2705Arg,P = 6.30 × 10(-9) in EAs and P = 2.98 × 10(-2) in AAs. Significant associations for FVIII :C were also observed with F8/TMLHE region SNP rs12557310 in EAs (P = 8.02 × 10(-10) ), with VWF rs1800380 in AAs (P = 5.62 × 10(-11) ), and with MAT1A rs2236568 in AAs (P51.69 × 10(-6) ). We replicated previously reported associations of FVIII :C and VWF :Ag with the ABO blood group, VWF rs1063856(Thr789Ala), rs216321 (Ala852Gln), and VWF rs2229446 (Arg2185Gln). Findings from this study expand our understanding of genetic influences for FVIII :C and VWF :Ag in both EAs and AAs.

The effect of a positively charged dodecyl trimethylammonium bromide (DTAB) surfactant on the aqueous solvation of coumarin 314 (C314), a solvatochromic molecule, at the air/water interface was investigated. Steady-state and femtosecond time-resolved second harmonic generation (SHG) spectroscopy was employed as a surface-selective tool to characterize the solvation of C314. The molecular orientation and the static and dynamic solvation of the probe C314 at the positively charged surfactant interface were found to be significantly different from that at the previously studied neat air/water interface and at a negatively charged surfactant air/water interface. Steady-state SHG spectra show a red shift and therefore increased local polarity at the positively charged interface (λmax = 436 nm) compared to the negatively charged interfaces (λmax = 431−432 nm) and the surfactant-free air/water interface (λmax = 423 nm), though less polar than bulk water (448 nm). Two diffusive dynamical time scales τ1 and τ2 of ...

Our objective is the development of robust methods for assessment of evidence for linkage of loci affecting a complex trait to a marker linkage group, using data on extended pedigrees. Using Markov chain Monte Carlo (MCMC) methods, it is possible to sample realizations from the distribution of gene identity by descent (ibd) patterns on a pedigree, conditional on observed data YM at multiple marker loci. Measures of gene ibdW which capture joint genome sharing in extended pedigrees often have unknown and highly skewed distributions, particularly when conditioned on marker data. MCMC provides a direct estimate of the distribution of such measures. Let W be the ibd measure from data YM, and W* the ibd measure from pseudo-data Y*M simulated with the same data availability and genetic marker model as the true data YM, but in the absence of linkage. Then measures of the difference in distributions of W and W* provide evidence for linkage. This approach extracts more information from the data YM than either comparison to the pedigree prior distribution of W or use of statistics that are expectations of W given the data YM. A small example is presented.

Objectives: To estimate the prevalence and incidence of hepatitis B virus (HBV) infection among patients attending three STD clinics in Pune, India, and to identify associated risk factors. Methods: Of the 2098 patients screened at STD clinics in Pune during 1996, 497, who returned for at least one follow up visit, were screened for various markers of HBV infection (HBsAg, anti-HBs, antiHBc), HIV antibody, and VDRL. Results: Of the 497 participants 3.6%, 26.5%, and 43.2% were positive for HBsAg, anti-HBs, and anti-HBc respectively. Tattooing (AOR 1.64, 95% CI 1.03 to 2.64) was found to be independently associated with presence of core antibody. Additionally, history of being in commercial sex work and history of a genital ulcer were independently associated with a positive anti-HBc antibody test (AOR 12.45, 95% CI 5.58 to 27.82 and AOR 1.70, 95% CI 1.09 to 2.66, respectively). 72 out of 497 (14.5%) participants were HIV positive at baseline. HIV-1 antibody positive patients were more likely to have a positive anti-HBc test (69.4% v 39.0%, p

Sickle cell trait may increase risk of venous thromboembolism, but this is not fully established.We sought to determine the association of sickle cell trait with deep vein thrombosis and pulmonary embolism.Middle-aged African Americans participating in a prospective, population-based cohort investigation, the Atherosclerosis Risk in Communities Study, were followed from 1987 through 2011 for incident hospitalized pulmonary embolism (n = 111) or isolated deep vein thrombosis (n = 138), verified by physician review of medical records. Sickle cell trait (heterozygosity for hemoglobin S, n = 268) was compared with no sickle cell trait (n = 3748).Over a median of 22 years of follow-up, 249 participants had an incident venous thromboembolism. The hazard ratio of venous thromboembolism was 1.50 (95% confidence interval [CI] 0.96-2.36) for participants with vs. without sickle cell trait, after adjustment for age, sex, ancestry, hormone replacement therapy (women), body mass index, diabetes, and estimated glomerular filtration rate. This hazard ratio was 2.05 (95% CI 1.12-3.76) for pulmonary embolism and 1.15 (95% CI 0.58-2.27) for deep vein thrombosis without pulmonary embolism.Sickle cell trait in African Americans carries a 2-fold increased risk of pulmonary embolism but does not elevate deep vein thrombosis risk. Because neonatal screening for sickle hemoglobin is being conducted in the United States, consideration should be paid to the increased pulmonary embolism risk of individuals with sickle cell trait.

Background: Genome-wide association studies (GWASs) have identified hundreds of genetic variants associated with complex diseases, but these variants appear to explain very little of the disease heritability. The typical single-locus association analysis in a GWAS fails to detect variants with small effect sizes and to capture higher-order interaction among these variants. Multilocus association analysis provides a powerful alternative by jointly modeling the variants within a gene or a pathway and by reducing the burden of multiple hypothesis testing in a GWAS. Methods: Here, we propose a powerful and flexible dimension reduction approach to model multilocus association. We use a Bayesian partitioning model which clusters SNPs according to their direction of association, models higher-order interactions using a flexible scoring scheme and uses posterior marginal probabilities to detect association between the SNP set and the disease. Results: We illustrate our method using extensive simulation studies and applying it to detect multilocus interaction in Atherosclerosis Risk in Communities (ARIC) GWAS with type 2 diabetes. Conclusion: We demonstrate that our approach has better power to detect multilocus interactions than several existing approaches. When applied to the ARIC study dataset with 9,328 individuals to study gene-based associations for type 2 diabetes, our method identified some novel variants not detected by conventional single-locus association analyses.

Objective— Tissue plasminogen activator (tPA), a serine protease, catalyzes the conversion of plasminogen to plasmin, the major enzyme responsible for endogenous fibrinolysis. In some populations, elevated plasma levels of tPA have been associated with myocardial infarction and other cardiovascular diseases. We conducted a meta-analysis of genome-wide association studies to identify novel correlates of circulating levels of tPA. Approach and Results— Fourteen cohort studies with tPA measures (N=26 929) contributed to the meta-analysis. Three loci were significantly associated with circulating tPA levels ( P −8 ). The first locus is on 6q24.3, with the lead single nucleotide polymorphism (SNP; rs9399599; P =2.9×10 −14 ) within STXBP5 . The second locus is on 8p11.21. The lead SNP (rs3136739; P =1.3×10 −9 ) is intronic to POLB and PLAT . We identified a nonsynonymous SNP (rs2020921) in modest linkage disequilibrium with rs3136739 ( r 2 =0.50) within exon 5 of PLAT ( P =2.0×10 −8 ). The third locus is on 12q24.33, with the lead SNP (rs7301826; P =1.0×10 −9 ) within intron 7 of STX2 . We further found evidence for the association of lead SNPs in STXBP5 and STX2 with expression levels of the respective transcripts. In in vitro cell studies, silencing STXBP5 decreased the release of tPA from vascular endothelial cells, whereas silencing STX2 increased the tPA release. Through an in silico lookup, we found no associations of the 3 lead SNPs with coronary artery disease or stroke. Conclusions— We identified 3 loci associated with circulating tPA levels, the PLAT region, STXBP5 , and STX2. Our functional studies implicate a novel role for STXBP5 and STX2 in regulating tPA release.

D-dimer, a fibrin degradation product, is related to risk of cardiovascular disease and venous thromboembolism. Genetic determinants of D-dimer are not well characterized; notably, few data have been reported for African American (AA), Asian, and Hispanic populations.We conducted a large-scale candidate gene association study to identify variants in genes associated with D-dimer levels in multi-ethnic populations. Four cohorts, comprising 6,848 European Americans (EAs), 2,192 AAs, 670 Asians, and 1,286 Hispanics in the National Heart, Lung, and Blood Institute Candidate Gene Association Resource consortium, were assembled. Approximately 50,000 genotyped single nucleotide polymorphisms (SNPs) in 2,000 cardiovascular disease gene loci were analyzed by linear regression, adjusting for age, sex, study site, and principal components in each cohort and ethnic group. Results across studies were combined within each ethnic group by meta-analysis.Twelve SNPs in coagulation factor V (F5) and 3 SNPs in the fibrinogen alpha chain (FGA) were significantly associated with D-dimer level in EAs with p2.0×10(-6). The signal for the most associated SNP in F5 (rs6025, factor V Leiden) was replicated in Hispanics (p=0.023), while that for the top functional SNP in FGA (rs6050) was replicated in AAs (p=0.006). No additional SNPs were significantly associated with D-dimer.Our study replicated previously reported associations of D-dimer with SNPs in F5 and FGA in EAs; we demonstrated replication of the association of D-dimer with FGA rs6050 in AAs and the factor V Leiden variant in Hispanics.

There is considerable interest in the role of vitamin D in human health. The best marker of vitamin D status is serum 25-hydroxyvitamin D [25(OH)D] [1]. An area that needs clarification is whether a low level of 25(OH)D may predispose to venous thromboembolism (VTE), as it may do for atherothrombotic diseases. Lindqvist et al suggested this possibility by showing in a cohort of Swedish women that sunbathers (with, presumably, extra sun-induced vitamin D had a 30% reduced VTE risk, and that population VTE risk was 50% lower in summer than winter (when vitamin D levels decrease) [2]. A large prospective study from Copenhagen reported an inverse association of seasonally-adjusted plasma 25(OH)D with VTE, with a relative risk of 1.28 (95% CI = 1.06, 1.53) for the lowest versus highest tertile of 25(OH)D [3]. However, the Tromso Study found no association of serum 25(OH)D with VTE, but was underpowered for low versus high 25(OH)D quartiles (relative risk 1.32, 95% CI = 0.78, 2.22) [4]. Administration of calcitriol, the active hormone form of vitamin D, was shown to reduce VTE occurrence in chronic kidney disease [5] and cancer patients [6]. Vitamin D receptor −/− mice, who display vitamin D deficiency, have increased thrombogenic activity [7]. 25(OH)D levels are seemingly deficient in 38% of black women, 18% of black men, 5% of white women, and 1% of white men, and they are optimal in only 2%, 4%, 29%, and 26%, respectively [8]. Grant speculated that a link between low vitamin D and VTE might explain an African American excess of VTE, compared with whites [9]. However, along with lower mean levels of 25(OH)D, recent evidence suggests that African Americans also have lower levels of vitamin D-binding protein and hence mean levels of bioavailable vitamin D similar to those in whites [10]. We tested prospectively in the Atherosclerosis Risk in Communities (ARIC) cohort whether 25(OH)D is inversely associated with VTE incidence and partly explains the African American excess of VTE in our sample [11].

We carried out a genome-wide association study (GWAS) for general cognitive ability (GCA) plus three other analyses of GWAS data that aggregate the effects of multiple single-nucleotide polymorphisms (SNPs) in various ways. Our multigenerational sample comprised 7,100 Caucasian participants, drawn from two longitudinal family studies, who had been assessed with an age-appropriate IQ test and had provided DNA samples passing quality screens. We conducted the GWAS across ∼ 2.5 million SNPs (both typed and imputed), using a generalized least-squares method appropriate for the different family structures present in our sample, and subsequently conducted gene-based association tests. We also conducted polygenic prediction analyses under five-fold cross-validation, using two different schemes of weighting SNPs. Using parametric bootstrapping, we assessed the performance of this prediction procedure under the null. Finally, we estimated the proportion of variance attributable to all genotyped SNPs as random effects with software GCTA. The study is limited chiefly by its power to detect realistic single-SNP or single-gene effects, none of which reached genome-wide significance, though some genomic inflation was evident from the GWAS. Unit SNP weights performed about as well as least-squares regression weights under cross-validation, but the performance of both increased as more SNPs were included in calculating the polygenic score. Estimates from GCTA were 35% of phenotypic variance at the recommended biological-relatedness ceiling. Taken together, our results concur with other recent studies: they support a substantial heritability of GCA, arising from a very large number of causal SNPs, each of very small effect. We place our study in the context of the literature-both contemporary and historical-and provide accessible explication of our statistical methods.

The absorption and fluorescence spectra of 3,3'-dioctadecyloxacarbocyanine [DiOC18(3)], a cationic oxacarbocyanine dye have been studied in aqueous and nonaqueous media containing egg phosphatidylcholine (PC) as well as in different solvents of diverse nature. The results show the evidence of complex formation of the dye in the ground and in the excited states with PC. The excited state interaction of the dye with PC suggests the electron transfer from PC to dye and this is supported by photovoltage generation in a photoelectrochemical cell consisting of dye and PC in aqueous medium. An attempt has been made to determine the polarity of the microenvironment of the dye in PC liposome or PC reverse micelle from the spectral studies of the dye in different solvents of known polarity.

Although twin, family, and adoption studies have shown that general cognitive ability (GCA) is substantially heritable, GWAS has not uncovered a genetic polymorphism replicably associated with this phenotype. However, most polymorphisms used in GWAS are common SNPs. The present study explores use of a different class of genetic variant, the copy-number variant (CNV), to predict GCA in a sample of 6,199 participants, combined from two longitudinal family studies. We aggregated low-frequency (

We report results from a genome wide association study (GWAS) of five quantitative indicators of behavioral disinhibition: nicotine, alcohol consumption, alcohol dependence, illicit drugs, and non-substance related behavioral disinhibition. The sample, consisting of 7,188 Caucasian individuals clustered in 2,300 nuclear families, was genotyped on over 520,000 SNP markers from Illumina’s Human 660W-Quad Array. Analysis of individual SNP associations revealed only one marker-component phenotype association, between rs1868152 and illicit drugs, with a p value below the standard genome-wide threshold of 5 × 10−8. Because we had analyzed five separate phenotypes, we do not consider this single association to be significant. However, we report 13 SNPs that were associated at p

Objectives: A gene-based genome-wide association study (GWAS) provides a powerful alternative to the traditional single single nucleotide polymorphism (SNP) association analysis due to its substantial reduction in the multiple testing burden and possible gain in power due to modeling multiple SNPs within a gene. A gene-based association analysis on multivariate traits is often of interest, but it imposes substantial analytical as well as computational challenges to implement it at a genome-wide level. Methods: We propose a rapid implementation of the multivariate multiple linear regression (RMMLR) approach in unrelated individuals as well as in families. Our approach allows for covariates. Moreover, the asymptotic distribution of the test statistic is not heavily influenced by the linkage disequilibrium (LD) among the SNPs and hence can be used efficiently to perform a gene-based GWAS. We have developed a corresponding R package to implement such multivariate gene-based GWAS with this RMMLR approach. Results: Through extensive simulation, we compared several approaches for both single and multivariate traits. Our RMMLR approach maintained a correct type I error level even for sets of SNPs in strong LD. It also demonstrated a substantial gain in power to detect a gene when it is associated with a subset of the traits. We also studied performances of the approaches on the Minnesota Center for Twin Family Research dataset. Conclusions: In our overall comparison, our RMMLR approach provides an efficient and powerful tool to perform a gene-based GWAS with single or multivariate traits and maintains the type I error appropriately.

Introduction: Activated partial thromboplastin time (aPTT) is commonly used to screen for coagulation factor deficiencies. Shorter aPTT is also a risk marker for incident and recurrent venous thromboembolism (VTE). Genetic factors influencing aPTT are not well understood. aPTT was associated with common genetic variants of coagulation factors V (F5), XI (F11), XII (F12), KNG1, HRG, and ABO in previously reported genome-wide association studies (GWAS) that were conducted in individuals of European ancestry; no data have been reported in other race groups. Hypothesis: The present study aimed to identify aPTT-related gene variants in European Americans (EAs) and African Americans (AAs). Methods: We conducted a large-scale candidate gene study for aPTT in 9,719 EAs and 2,799 AAs from the Atherosclerosis Risk in Communities (ARIC) study. Subjects on anticoagulants were excluded. Nearly 50,000 single nucleotide polymorphisms (SNPs) located in 2,100 candidate genes were genotyped by the Candidate gene Association Resource (CARe) gene chip. The association between each SNP and aPTT was assessed with an additive genetic model using linear regression adjusted for age, sex, and field center. We additionally adjusted for principal components in AAs to account for potential population stratification. P-value for significant threshold was set at 2x10-6 after accounting for multiple testing. Results: In EAs, fifty-five SNPs from F5, HRG, KNG1, F11, F12, and ABO genes exceeded the significant p-value threshold. The signals in HRG, KNG1, F11, F12, and ABO genes replicated the previously reported GWAS findings. The top variant in F5 identified in EAs was only weakly associated with the previously reported GWAS variant (rs2239852, p=1.89x10-08 and r2=0.02 with rs9332701 reported in the previously reported GWAS). In AAs, twenty-seven SNPs from the HRG, KNG1, F12, and ABO genes were significantly associated with aPTT. The top signals from the HRG (rs9898, p=1.19x10-27) and KNG1 genes ( rs710446 , p=8.41x10-42) replicated the previously reported signals in EAs with similar effect size and direction of association, but the top signals in the F12 and ABO genes were weakly associated with the previously reported variants in EAs (rs1801020 in F12: p=1.01x10-84 and r2=0.12 with rs2545801, and rs8176722 in ABO: p=1.62x10-29 and r2=0.26 with rs687621 , respectively). Conclusions: Our study replicated the previously reported associations of aPTT with HRG, KNG1, F11, F12, and ABO genes in EAs and with HRG and KNG1 in AAs. The signals from F5 identified in EAs and from F12 and ABO identified in AAs may represent new genetic variants for aPTT.

As part of the Genes, Environment and Development Initiative, the Minnesota Center for Twin and Family Research (MCTFR) undertook a genome-wide association study, which we describe here. A total of 8,405 research participants, clustered in four-member families, have been successfully genotyped on 527,829 single nucleotide polymorphism (SNP) markers using Illumina's Human660W-Quad array. Quality control screening of samples and markers as well as SNP imputation procedures are described. We also describe methods for ancestry control and how the familial clustering of the MCTFR sample can be accounted for in the analysis using a Rapid Feasible Generalized Least Squares algorithm. The rich longitudinal MCTFR assessments provide numerous opportunities for collaboration.

Background Greater public awareness of venous thromboembolism may be an important next step for optimizing venous thromboembolism prevention and treatment. "Lifetime risk" is an easily interpretable way of presenting risk information. Therefore, we sought to calculate the lifetime risk of venous thromboembolism (deep vein thrombosis or pulmonary embolism) using data from 2 large, prospective cohort studies: the Cardiovascular Health Study (CHS) and the Atherosclerosis Risk in Communities (ARIC) study. Methods We followed participants aged 45-64 years in ARIC (n = 14,185) and ≥65 in CHS (n = 5414) at baseline visits (1987-1989 in ARIC, 1989-1990 and 1992-1993 in CHS) for incident venous thromboembolism (n = 728 in ARIC through 2011 and n = 172 in CHS through 2001). We estimated lifetime risks and 95% confidence intervals of incident venous thromboembolism using a modified Kaplan-Meier method, accounting for the competing risk of death from other causes. Results At age 45 years, the remaining lifetime risk of venous thromboembolism in ARIC was 8.1% (95% confidence interval, 7.1-8.7). High-risk groups were African Americans (11.5% lifetime risk), those with obesity (10.9%), heterozygous for the factor V Leiden (17.1%), or with sickle cell trait or disease (18.2%). Lifetime risk estimates differed by cohort; these differences were explained by differences in time period of venous thromboembolism ascertainment. Conclusions At least 1 in 12 middle-aged adults will develop venous thromboembolism in their remaining lifetime. This estimate of lifetime risk may be useful to promote awareness of venous thromboembolism and guide decisions at both clinical and policy levels.

Allogeneic hematopoietic cell transplant (HCT) is associated with a high morbidity and mortality. Adhesion molecules play an important role in endothelial activation and initiation of inflammatory response. We hypothesized that single nucleotide polymorphisms (SNPs) in the endothelial molecules may contribute to heterogeneity in HCT outcomes. We evaluated the association of 4 SNPs in ICAM1 (rs5498), PECAM1 (rs668 and rs1131012) and SELL (rs2229569) genes with acute and chronic graft-versus-host disease (GvHD) and those experiencing transplant-related mortality (TRM) within 1 year among 425 allogeneic HCT recipient-donor pairs. Using a Fine and Gray proportional hazards model to evaluate the association between genetic variants and clinical outcomes, after adjustment for recipient age, race, diagnosis, disease status, gender mismatch, cytomegalovirus serostatus, gender, donor type, conditioning regimen and year of transplant, only rs5498 in the ICAM1 gene among both recipients and donors was associated with a decreased risk of TRM (P ≤ 0.02). None of the SNPs were associated with acute or chronic GvHD risk. These findings suggest that genetic variants in the vascular adhesion molecules may be used to identify patients at high risk for TRM.

Objective: Venous thromboembolism (VTE) is a common disease associated with a high mortality rate. A few variants in genes for factor V (FV Leiden), prothrombin, antithrombin, proteins C and S, and blood groups have been documented to confer VTE risk, but other genetic determinants for VTE likely exist. Methods: We performed an association study for VTE using a gene-centric approach in the Longitudinal Investigation of Thromboembolism Etiology (LITE) Study, which comprises the Atherosclerosis Risk in Communities (ARIC) and Cardiovascular Health Study (CHS) cohorts. A total of 280 VTE cases were ascertained in 9,827 participants of European ancestry in ARIC and 136 cases in 3,952 participants in CHS. The average follow-up time was 16 and 10 years in ARIC and CHS, respectively. Nearly 50,000 single nucleotide polymorphisms (SNPs) located in 2,100 candidate genes were genotyped by the Candidate gene Association Resource (CARe) gene chip. In each study, the association between each SNP and VTE was assessed with an additive genetic model using Cox proportional-hazards regression adjusted for age, sex, and field center. Data from ARIC and CHS were meta-analyzed using inverse-variance, fixed-effects models. SNPs with minor allele frequency < 1% or extreme p-values from Hardy-Weinberg disequilibrium test were excluded. Results: After adjusting for multiple testing, the thresholds for statistical significance and suggestive associations were set at 2.0x10−6 and 2.0x10−4, respectively. Four SNPs at the factor V gene locus exceeded the statistical significance threshold, with the top SNP being FV Leiden (rs6025, hazard ratio=3.01, p=9.7x10−17). Five SNPs showed suggestive associations with VTE (hazard ratios =1.29∼1.66), including those from the cholesteryl ester transfer protein (1 SNP), carboxypeptidase E (1 SNP), ABO (2 SNPs), and estrogen receptor 2 (1 SNP) genes. Conclusions: Among these genes, the FV Leiden variant and ABO are involved in coagulation and their associations with VTE have been previously confirmed. The other three genes have no known involvement in coagulation or hemostasis and thus, if replicated in future studies, may represent novel pathways.

Introduction: D-dimer, a marker of blood coagulation and fibrinolysis, is a widely-used biomarker of venous thromboembolism. While the coagulation factors V, III, II and fibrinogen alpha chain genes contain known genetic determinants of D-dimer, other determinants are not well characterized, especially in populations other than European Americans (EAs); no data have been reported for Asian American populations. Hypothesis: Large-scale candidate gene association analysis will identify D-dimer related gene variants in EA, African-American (AA), Chinese, and Hispanic populations. Methods: Four cohorts, comprised of 6,848 EA, 2,192 AAs, 670 Chinese, and 1,286 Hispanics from Cleveland Family Study, Cardiovascular Health Study, Framingham Heart Study, and Multi-Ethnic Study of Atherosclerosis in the Candidate Gene Association Resource (CARe) consortium, were included. The inverse normal transformed residual of D-dimer was used in genetic analysis. About 50,000 SNPs located in 2,100 inflammation and atherosclerosis-related genes were genotyped using the custom ITMAT-Broad-CARe (IBC) chip and analyzed by linear regression adjusted for age, sex, study site, and principal components to account for potential population stratification in each cohort and race group. Significance was set at p Results: Twelve SNPs in the coagulation factor V gene and 3 SNPs in the fibrinogen alpha chain genes exceeded the significant p-value threshold in EA. The top SNPs in these two regions were rs6025 (p-value=7.41x10-16) and rs13109457 (p-value=7.52x10-7), respectively. The signals for the remaining significant SNPs in these two regions were no longer statistically significant after adjusting for the known functional SNPs in each region (rs6025 and rs6050, respectively). The top signal at the coagulation factor V gene was replicated in Hispanics but not other race groups. Neither AA nor the other race groups replicated the signals for fibrinogen alpha chain gene. No additional variants in AA, Chinese, or Hispanics showed significant associations with D-dimer, but several SNPs in neuregulin 1 and Bruton agammaglobulinemia tyrosine kinase for AA and vanin 1 for Chinese were suggestively associated with D-dimer levels (p-value Conclusion: Our study replicated previously reported associations of D-dimer with coagulation factor V and fibrinogen alpha chain variants in EAs and identified coagulation factor V variation in Hispanics. Additional studies are required to confirm the suggestive findings in AAs, Chinese, and Hispanic populations.

In multi-locus association analysis, since some markers may not be associated with a trait, it seems attractive to use penalized regression with the capability of automatic variable selection. On the other hand, in spite of a rapidly growing body of literature on penalized regression, most focus on variable selection and outcome prediction, for which penalized methods are generally more effective than their non-penalized counterparts. However, for statistical inference, i.e. hypothesis testing and interval estimation, it is less clear how penalized methods would perform, or even how to best apply them, largely due to lack of studies on this topic. In our motivating data for a cohort of kidney transplant recipients, it is of primary interest to assess whether a group of genetic variants are associated with a binary clinical outcome, acute rejection at 6 months. In this paper, we study some technical issues and alternative implementations of hypothesis testing in Lasso penalized logistic regression, and compare their performance with each other and with several existing global tests, some of which are specifically designed as variance component tests for high-dimensional data. The most interesting, and perhaps surprising, conclusion of this study is that, for low to moderately high-dimensional data, statistical tests based on Lasso penalized regression are not necessarily more powerful than some existing global tests. In addition, in penalized regression, rather than building a test based on a single selected “best” model, combining multiple tests, each of which is built on a candidate model, might be more promising.

There has been an increasing interest in detecting gene-gene and gene-environment interactions in genetic association studies. A major statistical challenge is how to deal with a large number of parameters measuring possible interaction effects, which leads to reduced power of any statistical test due to a large number of degrees of freedom or high cost of adjustment for multiple testing. Hence, a popular idea is to first apply some dimension reduction techniques before testing, while another is to apply only statistical tests that are developed for and robust to high-dimensional data. To combine both ideas, we propose applying an adaptive sum of squared score (SSU) test and several other adaptive tests. These adaptive tests are extensions of the adaptive Neyman test [Fan, 1996], which was originally proposed for high-dimensional data, providing a simple and effective way for dimension reduction. On the other hand, the original SSU test coincides with a version of a test specifically developed for high-dimensional data. We apply these adaptive tests and their original nonadaptive versions to simulated data to detect interactions between two groups of SNPs (e.g. multiple SNPs in two candidate regions). We found that for sparse models (i.e. with only few non-zero interaction parameters), the adaptive SSU test and its close variant, an adaptive version of the weighted sum of squared score (SSUw) test, improved the power over their non-adaptive versions, and performed consistently well across various scenarios. The proposed adaptive tests are built in the general framework of regression analysis, and can thus be applied to various types of traits in the presence of covariates.

Linkage detection of a trait involves detecting regions of the genome that influence the trait. A wide variety of statistical models are currently employed for linkage analysis of quantitative traits. Many of these models are developed under some assumptions of the trait distributions. Violation of the assumptions about the trait generally affects the type I error and power for linkage detection. In this paper, we have proposed a trait-model-free approach for linkage analysis of a quantitative trait in general pedigrees. The conditional segregation of marker alleles given the trait is modeled using a latent-variable logistic model. A likelihood-ratio test is used to test for linkage under our model. The main applicability of this approach lies in the fact that it always provides correct type I error no matter what the trait distribution is and thus can be used for nonnormal traits or for selected samples. By means of simulation studies, we have compared the power of our proposed model with existing approaches for nonnormal traits. The performance of these methods was also studied on a real dataset. We have demonstrated the usefulness of our approach in terms of power and robustness for linkage detection of quantitative traits in general pedigrees.

Background— Plasma levels of coagulation factors VII (FVII), VIII (FVIII), and von Willebrand factor (vWF) influence risk of hemorrhage and thrombosis. We conducted genome-wide association studies to identify new loci associated with plasma levels. Methods and Results— The setting of the study included 5 community-based studies for discovery comprising 23 608 European-ancestry participants: Atherosclerosis Risk In Communities Study, Cardiovascular Health Study, British 1958 Birth Cohort, Framingham Heart Study, and Rotterdam Study. All subjects had genome-wide single-nucleotide polymorphism (SNP) scans and at least 1 phenotype measured: FVII activity/antigen, FVIII activity, and vWF antigen. Each study used its genotype data to impute to HapMap SNPs and independently conducted association analyses of hemostasis measures using an additive genetic model. Study findings were combined by meta-analysis. Replication was conducted in 7604 participants not in the discovery cohort. For FVII, 305 SNPs exceeded the genome-wide significance threshold of 5.0×10 −8 and comprised 5 loci on 5 chromosomes: 2p23 (smallest P value 6.2×10 −24 ), 4q25 (3.6×10 −12 ), 11q12 (2.0×10 −10 ), 13q34 (9.0×10 −259 ), and 20q11.2 (5.7×10 −37 ). Loci were within or near genes, including 4 new candidate genes and F7 (13q34). For vWF, 400 SNPs exceeded the threshold and marked 8 loci on 6 chromosomes: 6q24 (1.2×10 −22 ), 8p21 (1.3×10 −16 ), 9q34 (−324 ), 12p13 (1.7×10 −32 ), 12q23 (7.3×10 −10 ), 12q24.3 (3.8×10 −11 ), 14q32 (2.3×10 −10 ), and 19p13.2 (1.3×10 −9 ). All loci were within genes, including 6 new candidate genes, as well as ABO (9q34) and VWF (12p13). For FVIII, 5 loci were identified and overlapped vWF findings. Nine of the 10 new findings were replicated. Conclusions— New genetic associations were discovered outside previously known biological pathways and may point to novel prevention and treatment targets of hemostasis disorders.

Background There is a growing awareness that interaction between multiple genes play an important role in the risk of common, complex multi-factorial diseases. Many common diseases are affected by certain genotype combinations (associated with some genes and their interactions). The identification and characterization of these susceptibility genes and gene-gene interaction have been limited by small sample size and large number of potential interactions between genes. Several methods have been proposed to detect gene-gene interaction in a case control study. The penalized logistic regression (PLR), a variant of logistic regression with L 2 regularization, is a parametric approach to detect gene-gene interaction. On the other hand, the Multifactor Dimensionality Reduction (MDR) is a nonparametric and genetic model-free approach to detect genotype combinations associated with disease risk. Methods We compared the power of MDR and PLR for detecting two-way and three-way interactions in a case-control study through extensive simulations. We generated several interaction models with different magnitudes of interaction effect. For each model, we simulated 100 datasets, each with 200 cases and 200 controls and 20 SNPs. We considered a wide variety of models such as models with just main effects, models with only interaction effects or models with both main and interaction effects. We also compared the performance of MDR and PLR to detect gene-gene interaction associated with acute rejection(AR) in kidney transplant patients. Results In this paper, we have studied the power of MDR and PLR for detecting gene-gene interaction in a case-control study through extensive simulation. We have compared their performances for different two-way and three-way interaction models. We have studied the effect of different allele frequencies on these methods. We have also implemented their performance on a real dataset. As expected, none of these methods were consistently better for all data scenarios, but, generally MDR outperformed PLR for more complex models. The ROC analysis on the real dataset suggests that MDR outperforms PLR in detecting gene-gene interaction on the real dataset. Conclusion As one might expect, the relative success of each method is context dependent. This study demonstrates the strengths and weaknesses of the methods to detect gene-gene interaction.

Summary. Trait-model-free (or “allele-sharing”) approach to linkage analysis is a popular tool in genetic mapping of complex traits, because of the absence of explicit assumptions about the underlying mode of inheritance of the trait. The likelihood framework introduced by Kong and Cox (1997, American Journal of Human Genetics 61, 1179–1188) allows calculation of accurate p-values and LOD scores to test for linkage between a genomic region and a trait. Their method relies on the specification of a model for the trait-dependent segregation of marker alleles at a genomic region linked to the trait. Here we propose a new such model that is motivated by the desire to extract as much information as possible from extended pedigrees containing data from individuals related over several generations. However, our model is also applicable to smaller pedigrees, and has some attractive features compared with existing models (Kong and Cox, 1997), including the fact that it incorporates information on both affected and unaffected individuals. We illustrate the proposed model on simulated and real data, and compare its performance with the existing approach (Kong and Cox, 1997). The proposed approach is implemented in the program lm ibdtests within the framework of MORGAN 2.8 (http://www.stat.washington.edu/thompson/Genepi/MORGAN/Morgan.shtml).