Quantitative-Trait-Locus Analysis of Body-Mass Index and of Stature, by Combined Analysis of Genome Scans of Five Finnish Study Groups

In recent years, many genomewide screens have been performed, to identify novel loci predisposing to various complex diseases. Often, only a portion of the collected clinical data from the study subjects is used in the actual analysis of the trait, and much of the phenotypic data is ignored. With proper consent, these data could subsequently be used in studies of common quantitative traits influencing human biology, and such a reanalysis method would be further justified by the nonbiased ascertainment of study individuals. To make our point, we report here a quantitative-trait-locus (QTL) analysis of body-mass index (BMI) and stature (i.e., height), with genotypic data from genome scans of five Finnish study groups. The combined study group was composed of 614 individuals from 247 families. Five study groups were originally ascertained in genetic studies on hypertension, obesity, osteoarthritis, migraine, and familial combined hyperlipidemia. Most of the families are from the Finnish Twin Cohort, which represents a population-wide sample. In each of the five genome scans, approximately 350 evenly spaced markers were genotyped on 22 autosomes. In analyzing the genotype data by a variance-component method, we found, on chromosome 7pter (maximum multipoint LOD score of 2.91), evidence for QTLs affecting stature, and a second locus, with suggestive evidence for linkage to stature, was detected on chromosome 9q (maximum multipoint LOD score of 2.61). Encouragingly, the locus on chromosome 7 is supported by the data reported by Hirschhorn et al. (in this issue), who used a similar method. We found no evidence for QTLs affecting BMI.     

Quantitative founder-effect analysis of French Canadian families identifies specific loci contributing to metabolic phenotypes of hypertension

The Saguenay-Lac St-Jean population of Quebec is relatively isolated and has genealogical records dating to the 17th-century French founders. In 120 extended families with at least one sib pair affected with early-onset hypertension and/or dyslipidemia, we analyzed the genetic determinants of hypertension and related cardiovascular and metabolic conditions. Variance-components linkage analysis revealed 46 loci after 100,000 permutations. The most prominent clusters of overlapping quantitative-trait loci were on chromosomes 1 and 3, a finding supported by principal-components and bivariate analyses. These genetic determinants were further tested by classifying families by use of LOD score density analysis for each measured phenotype at every 5 cM. Our study showed the founder effect over several generations and classes of living individuals. This quantitative genealogical approach supports the notion of the ancestral causality of traits uniquely present and inherited in distinct family classes. With the founder effect, traits determined within population subsets are measurably and quantitatively transmitted through generational lineage, with a precise component contributing to phenotypic variance. These methods should accelerate the uncovering of causal haplotypes in complex diseases such as hypertension and metabolic syndrome.     

Quantitative variation in obesity-related traits and insulin precursors linked to the OB gene region on human chromosome 7.

Despite the evidence that human obesity has strong genetic determinants, efforts at identifying specific genes that influence human obesity have largely been unsuccessful. Using the sibship data obtained from 32 low income Mexican American pedigrees ascertained on a type II diabetic proband and a multipoint variance-components method, we tested for linkage between various obesity-related traits plus associated metabolic traits and 15 markers on human chromosome 7. We found evidence for linkage between markers in the OB gene region and various traits, as follows: D7S514 and extremity skinfolds (LOD = 3.1), human carboxypeptidase A1 (HCPA1) and 32,33-split proinsulin level (LOD = 4.2), and HCPA1 and proinsulin level (LOD = 3.2). A putative susceptibility locus linked to the marker D7S514 explained 56% of the total phenotypic variation in extremity skinfolds. Variation at the HCPA1 locus explained 64% of phenotypic variation in proinsulin level and approximately 73% of phenotypic variation in split proinsulin concentration, respectively. Weaker evidence for linkage to several other obesity-related traits (e.g., waist circumference, body-mass index, fat mass by bioimpedance, etc.) was observed for a genetic location, which is approximately 15 cM telomeric to OB. In conclusion, our study reveals that the OB region plays a significant role in determining the phenotypic variation of both insulin precursors and obesity-related traits, at least in Mexican Americans.     

Ordered-subsets linkage analysis detects novel Alzheimer disease loci on chromosomes 2q34 and 15q22

Alzheimer disease (AD) is a complex disorder characterized by a wide range, within and between families, of ages at onset of symptoms. Consideration of age at onset as a covariate in genetic-linkage studies may reduce genetic heterogeneity and increase statistical power. Ordered-subsets analysis includes continuous covariates in linkage analysis by rank ordering families by a covariate and summing LOD scores to find a subset giving a significantly increased LOD score relative to the overall sample. We have analyzed data from 336 markers in 437 multiplex (>/=2 sampled individuals with AD) families included in a recent genomic screen for AD loci. To identify genetic heterogeneity by age at onset, families were ordered by increasing and decreasing mean and minimum ages at onset. Chromosomewide significance of increases in the LOD score in subsets relative to the overall sample was assessed by permutation. A statistically significant increase in the nonparametric multipoint LOD score was observed on chromosome 2q34, with a peak LOD score of 3.2 at D2S2944 (P=.008) in 31 families with a minimum age at onset between 50 and 60 years. The LOD score in the chromosome 9p region previously linked to AD increased to 4.6 at D9S741 (P=.01) in 334 families with minimum age at onset between 60 and 75 years. LOD scores were also significantly increased on chromosome 15q22: a peak LOD score of 2.8 (P=.0004) was detected at D15S1507 (60 cM) in 38 families with minimum age at onset >/=79 years, and a peak LOD score of 3.1 (P=.0006) was obtained at D15S153 (62 cM) in 43 families with mean age at onset >80 years. Thirty-one families were contained in both 15q22 subsets, indicating that these results are likely detecting the same locus. There is little overlap in these subsets, underscoring the utility of age at onset as a marker of genetic heterogeneity. These results indicate that linkage to chromosome 9p is strongest in late-onset AD and that regions on chromosome 2q34 and 15q22 are linked to early-onset AD and very-late-onset AD, respectively.     

An Effective Method to Identify Heritable Components from Multivariate Phenotypes

Multivariate phenotypes may be characterized collectively by a variety of low level traits, such as in the diagnosis of a disease that relies on multiple disease indicators. Such multivariate phenotypes are often used in genetic association studies. If highly heritable components of a multivariate phenotype can be identified, it can maximize the likelihood of finding genetic associations. Existing methods for phenotype refinement perform unsupervised cluster analysis on low-level traits and hence do not assess heritability. Existing heritable component analytics either cannot utilize general pedigrees or have to estimate the entire covariance matrix of low-level traits from limited samples, which leads to inaccurate estimates and is often computationally prohibitive. It is also difficult for these methods to exclude fixed effects from other covariates such as age, sex and race, in order to identify truly heritable components. We propose to search for a combination of low-level traits and directly maximize the heritability of this combined trait. A quadratic optimization problem is thus derived where the objective function is formulated by decomposing the traditional maximum likelihood method for estimating the heritability of a quantitative trait. The proposed approach can generate linearly-combined traits of high heritability that has been corrected for the fixed effects of covariates. The effectiveness of the proposed approach is demonstrated in simulations and by a case study of cocaine dependence. Our approach was computationally efficient and derived traits of higher heritability than those by other methods. Additional association analysis with the derived cocaine-use trait identified genetic markers that were replicated in an independent sample, further confirming the utility and advantage of the proposed approach.     

Regression models for linkage heterogeneity applied to familial prostate cancer

Linkage heterogeneity frequently occurs for complex genetic diseases, and statistical methods must account for it to avoid severe loss in power to discover susceptibility genes. A common method to allow for only a fraction of linked pedigrees is to fit a mixture likelihood and then to test for linkage homogeneity, given linkage (admixture test), or to test for linkage while allowing for heterogeneity, using the heterogeneity LOD (HLOD) score. Furthermore, features of the families, such as mean age at diagnosis, may help to discriminate families that demonstrate linkage from those that do not. Pedigree features are often used to create homogeneous subsets, and LOD or HLOD scores are then computed within the subsets. However, this practice introduces several problems, including reduced power (which results from multiple testing and small sample sizes within subsets) and difficulty in interpretation of results. To address some of these limitations, we present a regression-based extension of the mixture likelihood for which pedigree features are used as covariates that determine the probability that a family is the linked type. Some advantages of this approach are that multiple covariates can be used (including quantitative covariates), covariates can be adjusted for each other, and interactions among covariates can be assessed. This new regression method is applied to linkage data for familial prostate cancer and provides new insights into the understanding of prostate cancer linkage heterogeneity.     

From QTL to QTN: Candidate Gene Set Approach and a Case Study in Porcine IGF1-FoxO Pathway

Unraveling the genetic background of economic traits is a major goal in modern animal genetics and breeding. Both candidate gene analysis and QTL mapping have previously been used for identifying genes and chromosome regions related to studied traits. However, most of these studies may be limited in their ability to fully consider how multiple genetic factors may influence a particular phenotype of interest. If possible, taking advantage of the combined effect of multiple genetic factors is expected to be more powerful than analyzing single sites, as the joint action of multiple loci within a gene or across multiple genes acting in the same gene set will likely have a greater influence on phenotypic variation. Thus, we proposed a pipeline of gene set analysis that utilized information from multiple loci to improve statistical power. We assessed the performance of this approach by both simulated and a real IGF1-FoxO pathway data set. The results showed that our new method can identify the association between genetic variation and phenotypic variation with higher statistical power and unravel the mechanisms of complex traits in a point of gene set. Additionally, the proposed pipeline is flexible to be extended to model complex genetic structures that include the interactions between different gene sets and between gene sets and environments.     

A genomewide scan for early-onset coronary artery disease in 438 families: the GENECARD Study

A family history of coronary artery disease (CAD), especially when the disease occurs at a young age, is a potent risk factor for CAD. DNA collection in families in which two or more siblings are affected at an early age allows identification of genetic factors for CAD by linkage analysis. We performed a genomewide scan in 1,168 individuals from 438 families, including 493 affected sibling pairs with documented onset of CAD before 51 years of age in men and before 56 years of age in women. We prospectively defined three phenotypic subsets of families: (1) acute coronary syndrome in two or more siblings; (2) absence of type 2 diabetes in all affected siblings; and (3) atherogenic dyslipidemia in any one sibling. Genotypes were analyzed for 395 microsatellite markers. Regions were defined as providing evidence for linkage if they provided parametric two-point LOD scores >1.5, together with nonparametric multipoint LOD scores >1.0. Regions on chromosomes 3q13 (multipoint LOD = 3.3; empirical P value <.001) and 5q31 (multipoint LOD = 1.4; empirical P value <.081) met these criteria in the entire data set, and regions on chromosomes 1q25, 3q13, 7p14, and 19p13 met these criteria in one or more of the subsets. Two regions, 3q13 and 1q25, met the criteria for genomewide significance. We have identified a region on chromosome 3q13 that is linked to early-onset CAD, as well as additional regions of interest that will require further analysis. These data provide initial areas of the human genome where further investigation may reveal susceptibility genes for early-onset CAD.     

Ordered stratification to reduce heterogeneity in linkage to diabetes-related quantitative traits

Phenotypic heterogeneity complicates detection of genomic loci predisposing to type 2 diabetes, potentially obscuring or unmasking specific loci. We conducted ordered-subsets linkage analyses (OSAs) for diabetes-related quantitative traits (fasting insulin and glucose, hemoglobin A1c (HbA1c), and 28-year-time-averaged fasting plasma glucose (FPG)) from 330 families of the Framingham Offspring Study. We calculated mean BMI, waist circumference (WC), and a diabetes "age-of-onset score" for each family. We constructed subsets by adding one family at a time in increasing (lean family to obese) or decreasing (obese to lean) adiposity order, or increasing or decreasing propensity to develop diabetes at a younger age, with the OSA LOD reported as the maximum LOD observed in any subset. Permutation P values tested the hypothesis that phenotypic ordering showed stronger linkage than random ordering. On chromosome 1, ordering by increasing family mean WC increased linkage to time-averaged FPG at 256 cM from LOD = 2.4 to 3.5 (permuted P = 0.02) and to HbA1c at 180 cM from LOD = 2.0 to 3.3 (P = 0.01). On chromosome 19, ordering by decreasing WC increased linkage to fasting insulin at 68 cM from LOD = 2.7 to 4.6 (P = 0.002), and ordering by decreasing propensity to develop diabetes at a young age increased linkage to fasting insulin at 73 cM from LOD = 2.7 to 4.0 (P = 0.046). We conclude that chromosomes 1 and 19 could harbor adiposity-interacting diabetes susceptibility genes. Such interactions might also influence trait-locus associations and may be useful to consider in diabetes genome-wide association studies.     

Genetic determination and linkage mapping of Plasmodium falciparum malaria related traits in Senegal

Plasmodium falciparum malaria episodes may vary considerably in their severity and clinical manifestations. There is good evidence that host genetic factors contribute to this variability. To date, most genetic studies aiming at the identification of these genes have used a case/control study design for severe malaria, exploring specific candidate genes. Here, we performed a family-based genetic study of falciparum malaria related phenotypes in two independent longitudinal survey cohorts, as a first step towards the identification of genes and mechanisms involved in the outcome of infection. We studied two Senegalese villages, Dielmo and Ndiop that differ in ethnicity, malaria transmission and endemicity. We performed genome-scan linkage analysis of several malaria-related phenotypes both during clinical attacks and asymptomatic infection. We show evidence for a strong genetic contribution to both the number of clinical falciparum malaria attacks and the asymptomatic parasite density. The asymptomatic parasite density showed linkage to chromosome 5q31 (LOD = 2.26, empirical p = 0.0014, Dielmo), confirming previous findings in other studies. Suggestive linkage values were also obtained at three additional chromosome regions: the number of clinical malaria attacks on chromosome 5p15 (LOD = 2.57, empirical p = 0.001, Dielmo) and 13q13 (LOD = 2.37, empirical p = 0.0014 Dielmo), and the maximum parasite density during asymptomatic infection on chromosome 12q21 (LOD = 3.1, empirical p<10(-4), Ndiop). While regions of linkage show little overlap with genes known to be involved in severe malaria, the four regions appear to overlap with regions linked to asthma or atopy related traits, suggesting that common immune related pathways may be involved.     

Individual-specific liability groups in genetic linkage, with applications to kindreds with Li-Fraumeni syndrome

In this report, we present a simple and powerful way to incorporate individual-specific liability classes into linkage analysis. The proposed method is applicable to both quantitative and qualitative traits. In linkage studies, we may have information about different covariates. Incorporation of these covariates along with the estimates of residual familial effects, age-at-onset effects, and susceptibility in the definition of liability classes can increase the power to detect genetic linkage. In this study, we show how one can form individual-specific liability classes and use these classes in standard linkage-analysis programs, such as the widely used LINKAGE package, to perform more powerful genetic linkage analysis. Our simulation study shows that this approach yields higher LOD scores and more-accurate estimates of the recombination fraction in the families showing linkage. The proposed method is also applied to kindreds collected, at the M. D. Anderson Cancer Center, through probands with childhood soft-tissue sarcoma. Confirmed germ-line mutations in the p53 tumor-suppressor gene have been identified in these families. Application of our method to these families yielded significantly higher LOD scores and more-accurate recombination fractions than did analysis that did not account for individual-specific covariate information.     

Genetic architecture of sexual dimorphism in a subdioecious plant with a proto-sex chromosome

The rise of sexual dimorphism is thought to coincide with the evolution of sex chromosomes. Yet because sex chromosomes in many species are ancient, we lack empirical evidence of the earliest stages of this transition. We use QTL analysis to examine the genetic architecture of sexual dimorphism in subdioecious octoploid Fragaria virginiana. We demonstrate that the region housing the male-function locus controls the majority of quantitative variation in proportion fruit set, confirming the existence of a proto-sex chromosome, and houses major QTL for eight additional sexually dimorphic traits, consistent with theory and data from animals and plants with more advanced sex chromosomes. We also detected autosomal QTL, demonstrating contributions to phenotypic variation in sexually dimorphic traits outside the sex-determining region. Moreover, for proportion seed set we found significant epistatic interactions between autosomal QTL and the male-function locus, indicating sex-limited QTL. We identified linked QTL reflecting trade-offs between male and female traits expected from theory and positive integration of male traits. These findings indicate the potential for the evolution of greater sexual dimorphism. Involvement of linkage groups homeologous to the proto-sex chromosome in these correlations reflects the polyploid origin of F. virginiana and raises the possibility that chromosomes in this homeologous group were predisposed to become the sex chromosome.     

Quantile-based permutation thresholds for quantitative trait Loci hotspots

Quantitative trait loci (QTL) hotspots (genomic locations affecting many traits) are a common feature in genetical genomics studies and are biologically interesting since they may harbor critical regulators. Therefore, statistical procedures to assess the significance of hotspots are of key importance. One approach, randomly allocating observed QTL across the genomic locations separately by trait, implicitly assumes all traits are uncorrelated. Recently, an empirical test for QTL hotspots was proposed on the basis of the number of traits that exceed a predetermined LOD value, such as the standard permutation LOD threshold. The permutation null distribution of the maximum number of traits across all genomic locations preserves the correlation structure among the phenotypes, avoiding the detection of spurious hotspots due to nongenetic correlation induced by uncontrolled environmental factors and unmeasured variables. However, by considering only the number of traits above a threshold, without accounting for the magnitude of the LOD scores, relevant information is lost. In particular, biologically interesting hotspots composed of a moderate to small number of traits with strong LOD scores may be neglected as nonsignificant. In this article we propose a quantile-based permutation approach that simultaneously accounts for the number and the LOD scores of traits within the hotspots. By considering a sliding scale of mapping thresholds, our method can assess the statistical significance of both small and large hotspots. Although the proposed approach can be applied to any type of heritable high-volume "omic" data set, we restrict our attention to expression (e)QTL analysis. We assess and compare the performances of these three methods in simulations and we illustrate how our approach can effectively assess the significance of moderate and small hotspots with strong LOD scores in a yeast expression data set.     

Differential genetic basis for pre-menopausal and post-menopausal salt-sensitive hypertension

Essential hypertension affects 75% of post-menopausal women in the United States causing greater cardiovascular complications compared with age-matched men and pre-menopausal women. Hormone replacement and current anti-hypertensive therapies do not correct this post-menopausal increased risk suggesting a distinct pathogenic framework. We investigated the hypothesis that distinct genetic determinants might underlie susceptibility to salt sensitive hypertension in pre-menopausal and post-menopausal states. To determine whether distinct genetic loci contribute to post-menopausal salt-sensitive hypertension, we performed a genome-wide scan for quantitative trait loci (QTLs) affecting blood pressure (BP) in 16-month old post-menopausal F2 (Dahl S×R)-intercross female rats characterized for blood pressure by radiotelemetry. Given identical environments and high salt challenge, post-menopausal BP levels were significantly higher than observed in pre-menopausal (post-menopausal versus pre-menopausal SBP, P<0.0001) and ovariectomized (post-menopausal versus ovariectomized SBP, P<0.001) F2-intercross female rats. We detected four significant to highly significant BP-QTLs (BP-pm1 on chromosome 13, LOD 3.78; BP-pm2 on chromosome 11, LOD 2.76; BP-pm3 on chromosome 2, LOD 2.61; BP-pm4 on chromosome 4, LOD 2.50) and two suggestive BP-QTLs (BP-pm5 on chromosome 15, LOD 2.37; BP-f1 on chromosome 5, LOD 1.65), four of which (BP-pm2, BP-pm3, BP-pm4, BP-pm5) were unique to this post-menopausal cohort. These data demonstrate distinct polygenic susceptibility underlying post-menopausal salt-sensitive hypertension providing a pathway towards the identification of mechanism-based therapy for post-menopausal hypertension and ensuing target-organ complications.     

Identification of genes for complex disease using longitudinal phenotypes

Using the simulated data set from Genetic Analysis Workshop 13, we explored the advantages of using longitudinal data in genetic analyses. The weighted average of the longitudinal data for each of seven quantitative phenotypes were computed and analyzed. Genome screen results were then compared for these longitudinal phenotypes and the results obtained using two cross-sectional designs: data collected near a single age (45 years) and data collected at a single time point. Significant linkage was obtained for nine regions (LOD scores ranging from 5.5 to 34.6) for six of the phenotypes. Using cross-sectional data, LOD scores were slightly lower for the same chromosomal regions, with two regions becoming nonsignificant and one additional region being identified. The magnitude of the LOD score was highly correlated with the heritability of each phenotype as well as the proportion of phenotypic variance due to that locus. There were no false-positive linkage results using the longitudinal data and three false-positive findings using the cross-sectional data. The three false positive results appear to be due to the kurtosis in the trait distribution, even after removing extreme outliers. Our analyses demonstrated that the use of simple longitudinal phenotypes was a powerful means to detect genes of major to moderate effect on trait variability. In only one instance was the power and heritability of the trait increased by using data from one examination. Power to detect linkage can be improved by identifying the most heritable phenotype, ensuring normality of the trait distribution and maximizing the information utilized through novel longitudinal designs for genetic analysis.     

Conditioning on subsets of the data: applications to ascertainment and other genetic problems.

I here consider the question of when to formulate a likelihood over the whole data set, as opposed to conditioning the likelihood on subsets of the data (i.e., joint vs. conditional likelihoods). I show that when certain conditions are met, these two likelihoods are guaranteed to be equivalent, and thus that it is generally preferable to condition on subsets, since that likelihood is mathematically and computationally simpler. However, I show that when these conditions are not met, conditioning on subsets of the data is equivalent to introducing additional df into our genetic model, df that we may not have been aware of. I discuss the implications of these facts for ascertainment corrections and other genetic problems.     

Fine mapping of the chromosome 12 late-onset Alzheimer disease locus: potential genetic and phenotypic heterogeneity

Apolipoprotein E (APOE) is the only confirmed susceptibility gene for late-onset Alzheimer disease (AD). In a recent genomic screen of 54 families with late-onset AD, we detected significant evidence for a second late-onset AD locus located on chromosome 12 between D12S373 and D12S390. Linkage to this region was strongest in 27 large families with at least one affected individual without an APOE-4 allele, suggesting that APOE and the chromosome 12 locus might have independent effects. We have since genotyped several additional markers across the region, to refine the linkage results. In analyzing these additional data, we have addressed the issue of heterogeneity in the data set by weighting results by clinical and neuropathologic features, sibship size, and APOE genotype. When considering all possible affected sib pairs (ASPs) per nuclear family, we obtained a peak maximum LOD score between D12S1057 and D12S1042. The magnitude and location of the maximum LOD score changed when different weighting schemes were used to control for the number of ASPs contributed by each nuclear family. Using the affected-relative-pair method implemented in GENEHUNTER-PLUS, we obtained a maximum LOD score between D12S398 and D12S1632, 25 cM from the original maximum LOD score. These results indicate that family size influences the location estimate for the chromosome 12 AD gene. The results of conditional linkage analysis by use of GENEHUNTER-PLUS indicated that evidence for linkage to chromosome 12 was stronger in families with affected individuals lacking an APOE-4 allele; much of this evidence came from families with affected individuals with neuropathologic diagnosis of dementia with Lewy bodies (DLB). Taken together, these results indicate that the chromosome 12 locus acts independently of APOE to increase the risk of late-onset familial AD and that it may be associated with the DLB variant of AD.     

Birt-Hogg-Dubé syndrome, a genodermatosis associated with spontaneous pneumothorax and kidney neoplasia, maps to chromosome 17p11.2

Birt-Hogg-Dubé syndrome (BHD), an inherited autosomal genodermatosis characterized by benign tumors of the hair follicle, has been associated with renal neoplasia, lung cysts, and spontaneous pneumothorax. To identify the BHD locus, we recruited families with cutaneous lesions and associated phenotypic features of the BHD syndrome. We performed a genomewide scan in one large kindred with BHD and, by linkage analysis, localized the gene locus to the pericentromeric region of chromosome 17p, with a LOD score of 4.98 at D17S740 (recombination fraction 0). Two-point linkage analysis of eight additional families with BHD produced a maximum LOD score of 16.06 at D17S2196. Haplotype analysis identified critical recombinants and defined the minimal region of nonrecombination as being within a <4-cM distance between D17S1857 and D17S805. One additional family, which had histologically proved fibrofolliculomas, did not show evidence of linkage to chromosome 17p, suggesting genetic heterogeneity for BHD. The BHD locus lies within chromosomal band 17p11.2, a genomic region that, because of the presence of low-copy-number repeat elements, is unstable and that is associated with a number of diseases. Identification of the gene for BHD may reveal a new genetic locus responsible for renal neoplasia and for lung and hair-follicle developmental defects.