Adjusting for confounding due to population admixture when estimating the effect of candidate genes on quantitative traits

Dosage transmission disequilibrium test (dTDT) for linkage and association detection

Both linkage and association studies have been successfully applied to identify disease susceptibility genes with genetic markers such as microsatellites and Single Nucleotide Polymorphisms (SNPs). As one of the traditional family-based studies, the Transmission/Disequilibrium Test (TDT) measures the over-transmission of an allele in a trio from its heterozygous parents to the affected offspring and can be potentially useful to identify genetic determinants for complex disorders. However, there is reduced information when complete trio information is unavailable. In this study, we developed a novel approach to "infer" the transmission of SNPs by combining both the linkage and association data, which uses microsatellite markers from families informative for linkage together with SNP markers from the offspring who are genotyped for both linkage and a Genome-Wide Association Study (GWAS). We generalized the traditional TDT to process these inferred dosage probabilities, which we name as the dosage-TDT (dTDT). For evaluation purpose, we developed a simulation procedure to assess its operating characteristics. We applied the dTDT to the simulated data and documented the power of the dTDT under a number of different realistic scenarios. Finally, we applied our methods to a family study of alcohol dependence (COGA) and performed individual genotyping on complete families for the top signals. One SNP (rs4903712 on chromosome 14) remained significant after correcting for multiple testing Methods developed in this study can be adapted to other platforms and will have widespread applicability in genomic research when case-control GWAS data are collected in families with existing linkage data.

Flexible semiparametric analysis of longitudinal genetic studies by reduced rank smoothing

In family-based longitudinal genetic studies, investigators collect repeated measurements on a trait that changes with time along with genetic markers. Since repeated measurements are nested within subjects and subjects are nested within families, both the subject-level and measurement-level correlations must be taken into account in the statistical analysis to achieve more accurate estimation. In such studies, the primary interests include to test for quantitative trait locus (QTL) effect, and to estimate age-specific QTL effect and residual polygenic heritability function. We propose flexible semiparametric models along with their statistical estimation and hypothesis testing procedures for longitudinal genetic designs. We employ penalized splines to estimate nonparametric functions in the models. We find that misspecifying the baseline function or the genetic effect function in a parametric analysis may lead to substantially inflated or highly conservative type I error rate on testing and large mean squared error on estimation. We apply the proposed approaches to examine age-specific effects of genetic variants reported in a recent genome-wide association study of blood pressure collected in the Framingham Heart Study.

Gene-environment interaction testing in family-based association studies with phenotypically ascertained samples: a causal inference approach

We propose a method for testing gene-environment (G × E) interactions on a complex trait in family-based studies in which a phenotypic ascertainment criterion has been imposed. This novel approach employs G-estimation, a semiparametric estimation technique from the causal inference literature, to avoid modeling of the association between the environmental exposure and the phenotype, to gain robustness against unmeasured confounding due to population substructure, and to acknowledge the ascertainment conditions. The proposed test allows for incomplete parental genotypes. It is compared by simulation studies to an analogous conditional likelihood-based approach and to the QBAT-I test, which also invokes the G-estimation principle but ignores ascertainment. We apply our approach to a study of chronic obstructive pulmonary disorder.

Unbiased and locally efficient estimation of genetic effect on quantitative trait in the presence of population admixture

Population admixture can be a confounding factor in genetic association studies. Family-based methods (Rabinowitz and Larid, 2000, Human Heredity 50, 211-223) have been proposed in both testing and estimation settings to adjust for this confounding, especially in case-only association studies. The family-based methods rely on conditioning on the observed parental genotypes or on the minimal sufficient statistic for the genetic model under the null hypothesis. In some cases, these methods do not capture all the available information due to the conditioning strategy being too stringent. General efficient methods to adjust for population admixture that use all the available information have been proposed (Rabinowitz, 2002, Journal of the American Statistical Association 92, 742-758). However these approaches may not be easy to implement in some situations. A previously developed easy-to-compute approach adjusts for admixture by adding supplemental covariates to linear models (Yang et al., 2000, Human Heredity 50, 227-233). Here is shown that this augmenting linear model with appropriate covariates strategy can be combined with the general efficient methods in Rabinowitz (2002) to provide computationally tractable and locally efficient adjustment. After deriving the optimal covariates, the adjusted analysis can be carried out using standard statistical software packages such as SAS or R. The proposed methods enjoy a local efficiency in a neighborhood of the true model. The simulation studies show that nontrivial efficiency gains can be obtained by using information not accessible to the methods that rely on conditioning on the minimal sufficient statistics. The approaches are illustrated through an analysis of the influence of apolipoprotein E (APOE) genotype on plasma low-density lipoprotein (LDL) concentration in children.

Association of the 98T ELAM-1 polymorphism with increased bleeding after cardiac surgery

OBJECTIVE

Hemorrhage continues to be a major problem after cardiac surgery despite the routine use of antifibrinolytic drugs, with striking inter-patient variability poorly explained by already known risk factors. The authors tested the hypothesis that genetic polymorphisms of inflammatory mediators and cellular adhesion molecules are associated with bleeding after cardiac surgery.

DESIGN

Prospective, observational study.

SETTING

Single, tertiary referral university heart center.

PARTICIPANTS

Adult patients undergoing aortocoronary surgery with cardiopulmonary bypass.

INTERVENTIONS

Patients (n = 759) had 10 mL of blood drawn preoperatively and genomic DNA isolated then genotyped for 17 polymorphisms in 7 candidate genes: tumor necrosis factor, interleukins 1beta and 6, interleukin 1 receptor antagonist, intercellular adhesion molecule-1 (ICAM-1), P-selectin and endothelial leucocyte adhesion molecule-1 (E-selectin). Multivariate analyses were used to relate clinical and genetic factors to bleeding and transfusion.

MEASUREMENTS AND MAIN RESULTS

The 98G/T polymorphism of the E-selectin gene was independently associated with bleeding after cardiac surgery (p = 0.002), after adjusting for significant clinical predictors (patient size and baseline hemoglobin concentration). There was a gene dose effect according to the number of minor alleles in the genotype; carriers of the minor allele bled 17% (GT) and 54% (TT) more than wild type (GG) genotypes, respectively (p = 0.01). Carriers of the minor allele also had longer activated partial thromboplastin times (p = 0.0023) and increased fresh frozen plasma transfusion (p = 0.03) compared with wild type.

CONCLUSIONS

The authors found a dose-related association between the 98T E-selectin polymorphism and bleeding after cardiac surgery, independent of and additive to standard clinical risk factors.

Methods and approaches in using secondary data sources to study race and ethnicity factors

Race and ethnicity are increasing used in cancer research to assess differences in cancer incidence and response to therapy. In this chapter, we discuss the measurement and methodologic issues that should be addressed to minimize bias and derive valid estimates when performing such assessments. These issues include 1) lack of national standards for race and ethnicity categories; 2) difficulty in comparing race and ethnic categories in longitudinal assessments; 3) broad categorization of race and ethnicity groups that do not provide adequate details for meaningful assessments; 4) inaccuracies in race and ethnicity data collection, and 5) confounding by socioeconomic and other factors. Recommendations for improving race and ethnicity data collection also are discussed.

Quantitative trait association in parent offspring trios: Extension of case/pseudocontrol method and comparison of prospective and retrospective approaches

The case/pseudocontrol method provides a convenient framework for family-based association analysis of case-parent trios, incorporating several previously proposed methods such as the transmission/disequilibrium test and log-linear modelling of parent-of-origin effects. The method allows genotype and haplotype analysis at an arbitrary number of linked and unlinked multiallelic loci, as well as modelling of more complex effects such as epistasis, parent-of-origin effects, maternal genotype and mother-child interaction effects, and gene-environment interactions. Here we extend the method for analysis of quantitative as opposed to dichotomous (e.g. disease) traits. The resulting method can be thought of as a retrospective approach, modelling genotype given trait value, in contrast to prospective approaches that model trait given genotype. Through simulations and analytical derivations, we examine the power and properties of our proposed approach, and compare it to several previously proposed single-locus methods for quantitative trait association analysis. We investigate the performance of the different methods when extended to allow analysis of haplotype, maternal genotype and parent-of-origin effects. With randomly ascertained families, with or without population stratification, the prospective approach (modeling trait value given genotype) is found to be generally most effective, although the retrospective approach has some advantages with regard to estimation and interpretability of parameter estimates when applied to selected samples.

Testing and estimating gene-environment interactions in family-based association studies

We propose robust and efficient tests and estimators for gene-environment/gene-drug interactions in family-based association studies in which haplotypes, dichotomous/quantitative phenotypes, and complex exposure/treatment variables are analyzed. Using causal inference methodology, we show that the tests and estimators are robust against unmeasured confounding due to population admixture and stratification, provided that Mendel's law of segregation holds and that the considered exposure/treatment variable is not affected by the candidate gene under study. We illustrate the practical relevance of our approach by an application to a chronic obstructive pulmonary disease study. The data analysis suggests a gene-environment interaction between a single nucleotide polymorphism in the Serpine2 gene and smoking status/pack-years of smoking. Simulation studies show that the proposed methodology is sufficiently powered for realistic sample sizes and that it provides valid tests and effect size estimators in the presence of admixture and stratification.

COGA phenotypes and linkages on chromosome 2

An initial linkage analysis of the alcoholism phenotype as defined by DSM-III-R criteria and alcoholism defined by DSM-IV criteria showed many, sometimes striking, inconsistencies. These inconsistencies are greatly reduced by making the definition of alcoholism more specific. We defined new phenotypes combining the alcoholism definitions and the latent variables, defining an individual as affected if that individual is alcoholic under one of the definitions (either DSM-III-R or DSM-IV), and indicated having a symptom defined by one of the latent variables. This was done for each of the two alcoholism definitions and five latent variables, selected from a canonical discriminant analyses indicating they formed significant groupings using the electrophysiological variables. We found that linkage analyses utilizing these latent variables were much more robust and consistent than the linkage results based on DSM-III-R or DSM-IV criteria for definition of alcoholism. We also performed linkage analyses on two first principal components derived phenotypes, one derived from the electrophysiological variables, and the other derived from the latent variables. A region on chromosome 2 at 250 cM was found to be linked to both of these derived phenotypes. Further examination of the SNPs in this region identified several haplotypes strongly associated with these derived phenotypes.

Enhanced APOE2 transmission rates in families with autistic probands

We have previously described linkage/association between reelin gene polymorphisms and autistic disorder. APOE also participates in the Reelin signaling pathway, by competitively antagonizing Reelin binding to APOE receptor 2 and to very-low-density lipoprotein receptors. The APOE2 protein variant displays the lowest receptor binding affinity compared with APOE3 and APOE4. In this study, we assess linkage/association between primary autism and APOE alleles in 223 complete trios, from 119 simplex Italian families and 44 simplex and 29 multiplex Caucasian-American families. Statistically significant disequilibrium favors the transmission of epsilon2 alleles to autistic offspring, over epsilon3 and epsilon4 (allele-wise transmission/disequilibrium test [TDT], chi2 = 6.16, 2 degrees of freedom [d.f.], P<0.05; genotype-wise TDT, chi2 = 10.68, 3 d.f., P<0.05). A novel epsilon3r allele was also discovered in an autistic child and his mother. Autistic patients do not differ significantly from unaffected siblings (allele-wise TDT comparing autistic patients versus unaffected sibs, chi2 = 1.83, 2 d.f., P<0.40, not significant). The major limitation of this study consists of our small sample size of trios including one unaffected sibling, currently not possessing the statistical power necessary to conclusively discriminate a specific association of epsilon2 with autism, from a distorted segregation pattern characterized by enhanced epsilon2 transmission rates both to affected and unaffected offspring. Our findings are thus compatible with either (a) pathogenetic contributions by epsilon2 alleles to autism spectrum vulnerability, requiring additional environmental and/or genetic factors to yield an autistic syndrome, and/or (b) a protective effect of epsilon2 alleles against the enhanced risk of miscarriage and infertility previously described among parents of autistic children.

PBAT: a comprehensive software package for genome-wide association analysis of complex family-based studies

The PBAT software package (v2.5) provides a unique set of tools for complex family-based association analysis at a genome-wide level. PBAT can handle nuclear families with missing parental genotypes, extended pedigrees with missing genotypic information, analysis of single nucleotide polymorphisms (SNPs), haplotype analysis, quantitative traits, multivariate/longitudinal data and time to onset phenotypes. The data analysis can be adjusted for covariates and gene/environment interactions. Haplotype-based features include sliding windows and the reconstruction of the haplotypes of the probands. PBAT's screening tools allow the user successfully to handle the multiple comparisons problem at a genome-wide level, even for 100,000 SNPs and more. Moreover, PBAT is computationally fast. A genome scan of 300,000 SNPs in 2,000 trios takes 4 central processing unit (CPU)-days. PBAT is available for Linux, Sun Solaris and Windows XP.

Detecting disease associations due to linkage disequilibrium using haplotype tags: a class of tests and the determinants of statistical power

In the 'indirect' method of detecting genetic associations between a trait and a DNA variant, we type several markers in a gene or chromosome region of linkage disequilibrium. If there is association between markers and the trait, we presume the existence of one or more causal polymorphisms in the region. In order to obtain a sufficiently dense set of markers it will almost always be necessary to use single nucleotide polymorphisms (SNPs). Although there is an emerging literature on methods for choosing an optimal set of 'haplotype tag SNPs' (htSNPs) to detect association between a genetic region and a trait, less attention has been given to the problem of how such studies should be analysed when completed, and how the initial data which was used to select the htSNPs should be incorporated into the analysis. This paper discusses this problem for both population- and family-based association studies. The role of the R2 measure of association between a causal locus and various methods of scoring of marker haplotypes is highlighted. In most cases, the simplest method of scoring (locus coding), which does not require phase resolution, is shown generally to be more powerful than scoring methods that include haplotype information. A new 'multi-locus TDT' is also proposed.

Candidate gene association analysis for a quantitative trait, using parent-offspring trios

With the increasing availability of genetic data, many studies of quantitative traits focus on hypotheses related to candidate genes, and also gene-environment (G x E) and gene-gene (G x G) interactions. In a population-based sample, estimates and tests of candidate gene effects can be biased by ethnic confounding, also known as population stratification bias. This paper demonstrates that even a modest degree of ethnic confounding can lead to unacceptably high type I error rates for tests of genetic effects. The parent-offspring trio design is reviewed, and several forms of the quantitative transmission disequilibrium test (QTDT) are summarized. A variation of the QTDT (QTDTM) is described that is based on a linear regression model with multiple intercepts, one per parental mating type. This and other models are expanded to allow testing of G x E and G x G interactions. A method for computing required sample sizes using direct computations is described. Sample size requirements for tests of genetic main effects and G x E and G x G interactions are compared across various QTDT approaches to infer their efficiencies relative to one another. The QTDTM is found to meet or exceed the efficiency of other QTDT approaches. For example, the QTDTM is approximately 3% more efficient than the QTDT of Rabinowitz ([1997] Hum. Hered. 47:342-350) for testing a genetic main effect, but can be as much as twice as efficient for testing G x E interaction, and three times more efficient for testing G x G interaction.

Genome-Wide Linkage Disequilibrium and Haplotype Maps

There is currently a broad effort to produce genome-wide high-density linkage disequilibrium (LD) maps with single nucleotide polymorphisms. The hope is that the resulting maps can be exploited to find genes that affect the onset and severity of at least some common human diseases. These maps may also be useful for identifying genes that affect drug response or the likelihood of drug toxicities. The goal of this review is to provide a broad overview of some of the key concerns motivating the design of a major international project called the International Haplotype Map Project. The process of map production requires the identification of very large numbers of polymorphic sites, implementation of facile, highly accurate and inexpensive genotyping production pipelines, and provision for public access to the genotype data. Great progress has been made recently in genotyping methods and these advances are allowing very large-scale data collection. A major goal of these efforts is to enable the selection of subsets of markers that capture useful genetic information in short genomic intervals, while optimally reducing the number of markers that must be genotyped. Standard measures of LD provide a starting point but may not fully capture the complexity of the information inherent in the data. Extremely dense genotype data in several broadly representative populations (European, Chinese, Japanese, and Yoruba) should yield important insights into the genetic structure of most genes. Further study is required to determine how broadly applicable the data will be to other population groups. Significant challenges lie ahead in determining the best methods for the selection of markers in disease/phenotype studies, large-scale genotyping, and analysis of the resulting genetic data.

Serotonin transporter gene promoter variants do not explain the hyperserotoninemia in autistic children

Autism is a biologically-heterogeneous disease. Distinct subgroups of autistic patients may be marked by intermediate phenotypes, such as elevated serotonin (5-HT) blood levels, potentially associated with different underlying disease mechanisms. This could lead to inconsistent genetic association results, such as those of prior studies on serotonin transporter (5-HTT) gene promoter variants and autistic disorder. Contributions of 5-HTT gene promoter alleles to 5-HT blood levels were thus investigated in 134 autistic patients and 291 first-degree relatives. Mean 5-HT blood levels are 11% higher in autistic patients carrying the L/L genotype, compared to patients with the S/S or S/L genotype; this trend is not observed in first-degree relatives. The probability of inheriting L or S alleles is significantly enhanced in patients with 5-HT blood levels above or below the mean, respectively (P < 0.05), but quantitative TDT analyses yield a non-significant trend (P = 0.10), as this polymorphism explains only 2.5% of the variance in 5-HT blood levels of autistic patients. In conclusion, 5-HTT gene promoter variants seemingly exert a small effect on 5-HT blood levels in autistic children, which largely does not account for hyperserotoninemia. Nonetheless, the inconsistent outcome of prior association studies could partly stem from a selection bias of hyper- or hypo-serotoninemic probands.

Family-based association studies

Over the past decade, attention has turned from positional cloning of Mendelian disease genes to the dissection of complex diseases. Both theoretical and empirical studies have shown that traditional linkage studies may be inferior in power compared to studies that directly utilize allele status. Case-control association studies, as an alternative, are subject to bias due to population stratification. As a compromise between linkage studies and case-control studies, family-based association designs have received great attention recently due to their potentially higher power to identify complex disease genes and their robustness in the presence of population substructure. In this review, we first describe the basic family-based association design involving one affected offspring with its two parents, all genotyped for a biallelic genetic marker. Extensions of the original transmission disequilibrium tests to multiallelic markers, families with multiple siblings, families with incomplete parental genotypes, and general pedigree structures are discussed. Further developments of statistical methods to study quantitative traits, to analyse genes on the X chromosome, to incorporate multiple tightly linked markers, to identify imprinting genes, and to detect gene-environment interactions are also reviewed. Finally, we discuss the implications of the completion of the Human Genome Project and the identification of hundreds of thousands of genetic polymorphisms on employing family-based association designs to search for complex disease genes.

Apolipoprotein epsilon2 allele is associated with an anti-atherogenic lipoprotein profile in children: The Columbia University BioMarkers Study

OBJECTIVE

We examined associations between allelic variation in the apo epsilon gene, which codes for apolipoprotein E, and plasma lipid levels in children.

MATERIALS AND METHODS

We analyzed genotype and fasting lipid levels, including lipid particle size by nuclear magnetic resonance spectroscopy, in 515 children from 297 families.

RESULTS

Children carrying the apo epsilon2 allele (1 or 2 epsilon2 alleles; n = 45) had higher mean high-density lipoprotein (HDL) cholesterol level (49.5 +/- 13.0 vs 42.4 +/- 8.9 mg/dL) and lower mean low-density lipoprotein (LDL) cholesterol level (82.2 +/- 48.6 vs 105.9 +/- 45.0 mg/dL) compared with apo epsilon3/epsilon3 children (n = 322). Mean HDL size was larger and mean level of the atheroprotective large HDL subpopulation was higher among apo epsilon2 carriers compared with epsilon3/epsilon3 children (9.5 +/- 0.4 vs 9.3 +/-.4 nm, and 32.8 +/- 9.9 vs 27.6 +/- 8.2 mg/dL). In multivariate models adjusting for age, sex, ethnicity, family history, body mass index, and fasting triglyceride level, the apo epsilon2 allele was independently predictive of higher levels of HDL cholesterol and the large HDL subpopulation and of lower level of LDL cholesterol.

CONCLUSION

The apo epsilon2 allele is associated with an anti-atherogenic lipid pattern in children.apolipoprotein epsilon, children, cholesterol.