FHS Offspring Cohort 100K Genome Scan Project Overview

The GMED website displays the results from the first high-density (116,000 SNPs/individual) genome-wide scan of the Framingham Heart Study (FHS) Offspring Cohort. 1320 participants in the FHS Offspring Cohort were genotyped as part of this project. The overall goal of the project has been to identify common genetic variants that contribute to the heritability of medically relevant traits such as body mass index, fasting blood glucose, plasma cholesterol, QT-interval and hypertension among others.

The purpose of the GMED (Genomic Medicine Database) website is to rapidly disseminate results from the genome scan and our analysis so that the large number of candidate associations can be tested for replication by other scientists. We have initiated replication studies in several cohorts and information on associations that replicate and on those that fail to replicate will be updated as the results of these studies become available. Please note that all associations in GMED are considered "candidate associations" pending replication in additional studies. We encourage all scientists to report information both for and against the associations presented here to us so that we can include this information on the website.

The study was performed by a team of scientists in the Genetics and Genomics department at Boston University School of Medicine (see Project Team). The GMED website was designed and implemented by Dr. Marc Lenburg and Mr. David Ulrich with input from others in the project team.

A detailed description of the analytical approaches used to identify candidate associations are specified for each association and described in detail under the Association Analysis tab.

All of the genotypes generated by our study have been returned to the National Heart, Lung and Blood Institute (NHLBI) and are freely available for any IRB-approved project. More information about requesting the 100K Scan genotypes and other Framingham Heart Study limited access data can be found at http://www.nhlbi.nih.gov/about/framingham/policies/index.htm

Mendelian versus Complex Disease

Over 1800 single gene or "Mendelian" or "single gene" disorders in humans are known. While obviously tragic for the affected individuals and their families, these diseases collectively have a very minor effect on overall public health because they are typically very rare. Because with Mendelian disorders every individual with the genetic variant that causes gets the disease, the genes responsible for these diseases have been relatively easy to identify. Thus, relatively small studies of disease transmission in rare affected families have been very effective at pinpointing these Mendelian disease genes. It should be noted that the study of Mendelian genetic diseases has contributed enormously to our collective understanding of human biology and disease.

In contrast, most heritable diseases and traits do not show this simple pattern of inheritance. These diseases are referred to as complex diseases because they are likely the result of the combined effects of many different genes and their complex interaction with environmental factors. The genes responsible for the complex pattern of inheritance seen in these types of diseases have been very hard to identify.

Why have complex disease genes been hard to find?

Despite excellent evidence for their existence, the common variants that predispose to common diseases are largely unknown as of August 2006. One reason is that common diseases arise from the interaction of several independently acting risk genes and environmental risks. This means that the effect of any one gene is probably small. To identify small effects, one must examine large numbers of individuals. Additionally, an individual inheriting a complex disease gene has only an elevated risk of disease. This means that some individuals carrying the disease predisposing allele will not be affected.

Another reason complex disease genes have been hard to find is that until now the search for genetic associations has focused, by necessity, on the tiny fraction of the genome that harbors genes that are suspected to play a role in disease. Such "candidate gene" studies have met with limited success, but so far have not identified many truly common variants that increase disease predisposition. To find the common variants that increase disease risk, untargeted scans that associate common genetic variants with disease in the unexplored 99% of the genome are needed.

Such studies can only now be accomplished because the necessary foundation has been laid. This foundation consists of several key elements [Hirschhorn & Daly, 2005]:

1. the human genome sequence (the result of massive public and private efforts)
2. discovery of nearly all 11 million common genetic variants (SNP consortium)
3. a dense map of genetic variation in several world populations (the HapMap)
4. technical advances in genotyping that have made large scale studies both faster and cost effective

Candidate Gene Versus Genome-Wide Association Studies

Limitations of the Candidate Gene Approach. Genetic association studies seek to identify complex disease loci through the overrepresentation of polymorphic sites, such as SNPs, among groups of diseased versus healthy individuals. This can be accomplished in either case-control or population-based formats. While many common genetic variants have been reported to be associated with disease phenotypes, remarkably few of these associations are robust to replication. Recent meta-analysis of reported associations showed that over half do not replicate in follow up studies and >95% are not consistently replicated [Hirschhorn et al., 2002], owing to the confounding effects of population stratification, sample ascertainment and other factors. For example, the lack of consistent replication makes it challenging to interpret the more than 70 reported associations between body mass index (weight in kilograms/height in meters squared) and common genetic variants identified through candidate gene studies [Perusse et al., 2005].

Untargeted "Genome Scans" to Associate Genetic Variation with Disease. Historically, association studies were only feasible in a candidate gene approach in which several hundred candidate gene regions were typed for polymorphic markers. In principle, however, it is possible to perform an unbiased "genome-scan" in which no prior assumptions are made as to likely candidate genes. The use of genome-wide scans has been restricted due to the need to have hundreds of thousands of mapped common SNPs, the cost of such dense SNP genotyping, and statistical approaches to handle the multiple-testing problem that arises when hundreds of thousands of possible associations are examined. Great progress has been made on all fronts in the last few years with ~11 million SNPs now in the dbSNP database, the cost of genotyping at roughly 0.1 cents per SNP [Hirschhorn & Daly, 2005] and the development of methods to handle the multiple testing problem such as PBAT [Van Steen et al., 2005] and staged genotyping (Daly and Hirschhorn, 2005). Indeed, the first generation of genome scan projects is now underway. We have recently reported an obesity-predisposing variant that we discovered in an untargeted scan of the Framingham Heart Study (FHS) Offspring Cohort [Herbert et al., 2006].

Human Genetic Variation

The speed and success of the human genome project, the SNP consortium [Kruglyak and Nickerson, 2001] and the HapMap project [Altshuler et al., 2005, Gabriel et al., 2002, Gibbs, 2003]) have served to lay the foundation for the next generation of efforts to map complex disease genes and the quantitative trait loci (QTLs) that may be preclinical indicators of pending disease.

Understanding the nature of human genetic variation has profound consequences for studies of complex disease. The fact that over 90% of the major class of variant, single nucleotide polymorphisms (SNPs), are polymorphic in all populations indicates that a great deal of human genetic variation is ancient, arising mostly in the many generations preceding migration out of Africa [Altshuler et al., 2005].

The result is that humans around the globe are surprisingly closely related. Indeed, if any two people are compared in DNA sequence they differ at only 1 in 1000 bases. If we examine the sites (or SNPs) that differences between these two people in other people unrelated people, a remarkable result is that the same sites are found to vary.

If many of the variants that predispose to complex diseases such as heart disease, obesity, cancer and asthma are also ancient, as postulated by the common disease/common variant hypothesis [Chakravarti, 1999, Reich and Lander, 2001], then a map of common variants should provide a useful tool to pinpoint the location of complex disease genes. Our recent finding of an obesity-predisposing variant present in 10% of people worldwide [Herbert et al., 2006] supports the common disease/common variant idea.

With the SNP consortium's effort to identify common variants (minor allele frequency > 1%) nearing its end, the majority of the estimated 11 million such variants have now been identified [Reich et al., 2003]. A large number of these variants have been examined for patterns of LD [Gabriel et al., 2002] in several ethnic populations by the HapMap project and the results are publicly available [Altshuler et al., 2005]. Thus, with this density of well-characterized markers, human geneticists are poised for the first time to rapidly find the common variants that contribute to common diseases.

references

Herbert, A., N.P. Gerry, M.B. McQueen, I.M. Heid, A. Pfeufer, T. Illig, H.E. Wichmann, T. Meitinger, D. Hunter, F.B. Hu, G. Colditz, A. Hinney, J. Hebebrand, K. Koberwitz, X. Zhu, R. Cooper, K. Ardlie, H. Lyon, J.N. Hirschhorn, N.M. Laird, M.E. Lenburg, C. Lange, and M.F. Christman. A common genetic variant is associated with adult and childhood obesity. Science. 2006. 312(5771): p. 279-83.

Altshuler, D., Brooks, L. D., Chakravarti, A., Collins, F. S., Daly, M. J., and Donnelly, P. (2005). A haplotype map of the human genome. Nature 437, 1299-1320.

Hirschhorn, J. N., and Daly, M. J. (2005). Genome-wide association studies for common diseases and complex traits. Nat Rev Genet 6, 95-108.

Perusse, L., Rankinen, T., Zuberi, A., Chagnon, Y. C., Weisnagel, S. J., Argyropoulos, G., Walts, B., Snyder, E. E., and Bouchard, C. (2005). The human obesity gene map: the 2004 update. Obes Res 13, 381-490.

Van Steen K, McQueen MB, Herbert A, Raby B, Lyon H, Demeo DL, Murphy A, Su J, Datta S, Rosenow C, et al: Genomic screening and replication using the same data set in family-based association testing. Nat Genet 2005, 37:683-691.

Gibbs, R. A. e. a. (2003). The International HapMap Project. Nature 426, 789-796.

Reich, D. E., Gabriel, S. B., and Altshuler, D. (2003). Quality and completeness of SNP databases. Nat Genet 33, 457-458.

Gabriel, S. B., Schaffner, S. F., Nguyen, H., Moore, J. M., Roy, J., Blumenstiel, B., Higgins, J., DeFelice, M., Lochner, A., Faggart, M., et al. (2002). The structure of haplotype blocks in the human genome. Science 296, 2225-2229.

Hirschhorn, J. N., Lohmueller, K., Byrne, E., and Hirschhorn, K. (2002). A comprehensive review of genetic association studies. Genet Med 4, 45-61.

Kruglyak L, Nickerson DA: Variation is the spice of life. Nat Genet 2001, 27:234-236.

Reich, D. E., and Lander, E. S. (2001). On the allelic spectrum of human disease. Trends Genet 17, 502-510.

Chakravarti, A. (1999). Population genetics--making sense out of sequence. Nat Genet 21, 56-60.