Correspondence *Center for Health Research, 100 North Academy Lane, Danville, PA 17822, USA

Email kaskelding@geisinger.edu

Introduction

One person, trillions of cells, 23 pairs of chromosomes, 30,000 genes, 3,164,700,000 base pairs—the complexity is astounding. From taking months or years to identify and sequence a single gene, we have made leaps and bounds and now use DNA chips to identify up to 500,000 genetic markers in just a few days. Despite the dramatic improvement in processing time, the genetic variations responsible for many of the common cardiovascular diseases have remained elusive. The Haplotype Genetic Map (HapMap) has recently been published, however, adding to researchers' armamentarium by providing an invaluable resource to increase the efficiency, decrease the cost and increase the speed of our genetic discoveries in the laboratory.^{1, 2, 3, 4} What does this development mean for our cardiovascular practice? Will the HapMap change the way we prescribe and use medications and how we choose devices in the catheterization laboratory, as well as improve patient outcome? The answer to all these questions is probably yes, and these changes will be seen in the near future with the advent of the genetic revolution. Genomics is the evaluation of genes as a dynamic system, in which genes interact to influence biologic pathways, networks and physiology. To take full advantage of the HapMap, we need to learn the language, understand the implications and develop unambiguous phenotypes for common clinical disorders.⁵

Practicing cardiologists regularly face clinical scenarios that are likely to have a strong genetic origin, but for which the molecular basis has not yet been determined. For example, patients who undergo successful angioplasty and stent implantation, performed by the best interventional cardiologists with diligent attention paid to mechanical and pharmacologic details, continue to endure the risk of in-stent restenosis and subacute thrombosis. Candidate gene testing of the genes most likely to be involved has provided some clues to the etiology of in-stent restenosis and subacute thrombosis, but has provided a less-than-complete picture of the multivariable process. Phenotypes are, by definition, the expression of specific traits and are based on genetic and environmental influences. Biological variables, such as platelet reactivity, pharmacologic resistance and cellular proliferation, that determine complex disease phenotypes, such as subacute thrombosis and restenosis, are intertwined with environmental variables, such as diet, lifestyle, age, sex and risk-factor therapies. This interplay makes it difficult to identify causative genes.^{6, 7, 8, 9, 10, 11, 12, 13} One hypothesis for the presence of such common but complex disease phenotypes is that they are caused by multiple common genetic variants working in concert with environmental factors.^{3, 14, 15, 16, 17, 18, 19} Only by elucidating the complex interactions of such genetic variants, which are weak individually but together confer a deleterious phenotype, will we begin to understand and treat or prevent effectively processes such as in-stent restenosis and subacute thrombosis.²⁰

The identification of these common variants is precisely where the HapMap promises to be a key resource. Most breakthroughs in genetics have been made through linkage studies performed in rarer, more severe, single-gene Mendelian diseases. In common complex diseases a number of genetic variants are expected to underlie the clinical phenotype. The synergistic effects of multiple genetic variants, along with environmental factors that act synergistically to develop the phenotype, make it more difficult to identify the actual causative genetic variants. Classic case–control association studies using the HapMap could provide the experimental design to circumvent this difficulty.^{21, 22} In this Review, we examine the HapMap and provide insight into the relevance of this work to cardiovascular practice, clinical research in cardiovascular disease and future discoveries in diagnostic and therapeutic modalities.

The genomics revolution

For many of us, the last formal genetics we studied was in medical school. The vernacular then included DNA, RNA and chromosomes but had little or no mention of the current genetic terminology of single nucleotide polymorphisms (SNPs; pronounced 'snips'), haplotypes and chips. The focus was on monogenic disorders, which are controlled by a single gene, and are associated with a distinct pattern of inheritance. In 2001, a milestone was reached with the publication of the first draft of the Human Genome Project. This development brought industry and academic research together to provide a complete view of the human genome (i.e. all of the DNA in each nucleated cell of an organism, including the DNA that encodes for genes as well as the interspersed regions between genes).²³ Before 2001, the focus was on amalgamation and construction—all the sequenced portions from all the laboratories involved were put together as puzzle pieces to construct the complete genome.

One important finding of the Human Genome Project was that approximately 99.9% of the genetic code is identical between individuals. When the DNA sequences of two humans are compared, only approximately 1 in every 1,200 bases, on average, is different. For example, one person might have an adenine at a given location, while another person might have a guanine. The order of these bases, however, is of great importance as it underlies the genetic diversity we see among humans and nonhuman organisms. In fact, by comparing genomes across species we can identify which parts of the DNA have remained unchanged (are 'highly conserved') through time. This highly conserved DNA can provide insight into important gene regulatory regions.^{24, 25} Highly conserved segments found in noncoding regions—originally thought to be useless, but which at times carry the information for gene regulation—can affect the expression of the corresponding RNA or protein. With other types of variations, one person could have extra bases (insertions) or fewer bases (deletions).

Differences in single bases are by far the most common and are known as SNPs. Thus far, at least 3,500,000 SNPs have been identified and approximately 10 million SNPs are estimated to be present in the human genome.^{26, 27} The relationship between disease phenotypes and SNPs that occur in genes with known function, such as the angiotensin-converting-enzyme polymorphism D/I and its relationship to in-stent restenosis, have been tested with discouraging results.^{8, 9, 12, 28, 29, 30, 31} Candidate genes with known SNPs are selected because they are identified or suspected to be involved in the disease process on the basis of existing knowledge, such as animal models or cell-culture studies. Common diseases, however, such as in-stent restenosis and myocardial infarction, are likely to result from complex gene–gene and gene–environment interactions that might not readily portend their phenotype.^{6, 14, 16, 17, 19, 32, 33, 34, 35}

What is the HapMap?

The HapMap is a map, resource, tool and a catalogue of common haplotypes. A haplotype is a group of genetic variants, or genotypes closely linked on one chromosome and inherited as a unit. Haplotypes can reside within genes and specify alleles, alternative forms of a gene, which can result in quantitatively or qualitatively different gene products and, consequently, different phenotypes. A single allele is inherited from each parent. A person is homozygous if the allele on each chromosome is identical, and is heterozygous if the alleles are different. The HapMap catalogues the locations of common haplotypes throughout the genome, details what these variants are, and describes how they are distributed across populations. In contrast to the Human Genome Project, which generated a 'reference' sequence of all chromosomes from only a few individuals, the HapMap is the logical next phase to begin the characterization of the human genome by determining the extent of genetic variability.^{23, 24, 36}

The HapMap project was launched in 2002 through an international consortium^{1, 27} and includes four population samples: 30 trios (a trio is a mother–father–child group and is a classic genetic collection) of North American descent (from Utah with ancestry from northern and western Europe; CEPH/Utah), 30 trios of Nigerian descent (from the Yoruba in Ibadan, Nigeria), 48 unrelated Japanese (from Tokyo, Japan) and 48 unrelated Chinese individuals (Han Chinese from Beijing).^{1, 3} The study of this important but limited number of samples provides insight into 'common' genetic variability and will need to be extended into further populations to become more widely generalizable.

In the first phase of the project, a map was developed that identified evenly spaced SNPs (spaced approximately every 5 kb) that had a minor allele frequency greater than 5%. SNPs that are evenly spaced throughout the genetic map allow analysis of linkage disequilibrium (LD) patterns and identification of haplotype blocks. LD is a somewhat counterintuitive term that refers to the extent to which alleles within a population are found together. In a large population over time, alleles achieve equilibrium due to the recombination by which they are randomly inherited. If alleles are in disequilibrium, this signifies that they are not randomly inherited but are inherited together as a unit, generally referred to as a haplotype block. Benefits associated with the identification of haplotype blocks rather than analysis of individual SNPs include the ability to reduce the amount of tests needed to find a genomic area of interest and the ability to use a marker SNP to find a disease-causing SNP. These SNPs are usually inherited along with adjacent SNPs on the same chromosome or haplotype block. As a result of the coinheritance of many of the SNPs that are in linkage disequilibrium and part of a haplotype block, a variant base unique to the set of SNPs can often be used to serve as a marker or 'tag' for the entire SNP group (Figure 1). For example, when a genotype at one SNP is known and the association between this SNP and another is high (i.e. they are in LD), we can, therefore, predict the genotype at the second SNP without performing genotyping. As many SNPs could be in LD, testing for the presence of only one of these allows for a decrease in the number of SNPs used to identify the genetic cause of disease.^{23, 36} This technique, therefore, decreases both time and cost without compromising the information gained. LD across the human genome has been characterized using the HapMap data.^{37, 38, 39} It is low near the ends of each chromosome, which indicates that there is much recombination in these areas. By contrast, LD increases near the centromeres. Interestingly, regions of strong LD tend to have fewer guanine and cytosine bases and do not have as many genetic polymorphisms as other regions. Some classes of genes, such those involved in immune response and sensory perception, are typically located in regions of low LD; while other classes of genes, including those involved in DNA and RNA metabolism, response to DNA damage, and the cell cycle, are largely located in regions of high LD.

Figure 1 SNPs and haplotype blocks

(A) SNPs. Shown is a short stretch of DNA from four versions of the same chromosome region in different people. Most of the DNA sequence is identical in these chromosomes, but three bases are shown where variation occurs. Each SNP has two possible alleles; the first SNP in panel A has the alleles cytosine and thymine. (B) Haplotypes. A haplotype is made up of a particular combination of alleles at nearby SNPs. Shown here are the observed genotypes for 20 SNPs that extend across 6,000 bases of DNA. Only the variable bases are shown, which include the three SNPs that are shown in panel A. For this region, most of the chromosomes in a population survey turn out to have haplotypes 1–4. (C) Tag SNPs. Genotyping just the three tag SNPs out of the 20 SNPs is sufficient to identify these four haplotypes uniquely. For instance, if a particular chromosome has the pattern A–T–C at these three tag SNPs, this pattern matches the pattern determined for haplotype 1. Abbreviation: SNP, single nucleotide polymorphism. Reproduced with permission from The International HapMap Consortium (2003) The International HapMap Project. Nature426: 789–796.

Full figure and legend (49K)Figures & Tables indexDownload Power Point slide (253K)

Evolutionary insights have also been made. For example, data from the HapMap have been used to support the hypothesis that as domestication of dairy animals increased milk consumption, variations in lactase gene expression that enabled humans to retain the ability to digest lactose in adulthood were selectively favored and quickly rose to high frequency in human populations.^{40, 41}

In addition, identification of tag SNPs associated with a given disease allows the HapMap to be used to quickly find candidate genes that can be studied further, such as in the development of animal models (e.g. knockout mice) in which diagnostic and therapeutic techniques can be tested and refined in preparation for phase I human clinical trials.

Analysis of the HapMap data has also been used to determine estimates of genetic population structure.⁴¹ As expected, a high similarity was observed between the Han Chinese cohort from Beijing and the Japanese cohort from Tokyo. Interestingly, discrete regions of dissimilarity were found between these two populations, as well as between the Yoruba cohort from Nigeria and Whites of European descent, which could be explored to identify potential candidate genes underlying the differences in cardiovascular disease phenotypes among various ethnic populations.^{42, 43}

The HapMap and cardiovascular research

Currently, the only haplotyping that has been adopted into routine clinical practice is human leukocyte antigen matching in organ transplantation. Matching the donor and recipient on the basis of this test can help predict which patients will accept and which might reject the implanted organ, and is vital to the success of organ transplantation.⁴⁴ Although no genome-wide haplotype analysis has been reported so far in cardiovascular medicine, a number of studies have demonstrated the contribution that haplotype identification can make to the assignment of risk in a number of patient subgroups. For example, susceptibility to sudden cardiac death because of improper modulation of cardiac conduction, which causes long QT syndrome, was found to be associated with the voltage-gated sodium channel type V (SCN5A) promoter haplotype, HapB, and displayed dose response.¹⁵ For example, if the two alleles were A and B, patients could be AA, AB or BB at this locus. If A were a dominant deleterious allele, individuals with AA would have higher susceptibility than those with AB, and those with BB would not be susceptible. By contrast, if A were a recessive trait, the phenotype would only be seen with the AA genotype, and AB individuals would be protected due to the dominant and safe B allele. Investigation into genetic variants of proprotein convertase subtilisin/kexin type 9 (PCSK9), known to be associated with familial hypercholesterolemia, found the haplotype 3 variant was independently associated with LDL cholesterol level in patients with polygenic hypercholesterolemia (odds ratio [OR] 2.36, 95% CI 1.90–4.32; P = 0.005).¹⁴

Acute myocardial infarction risk has been associated with haplotypes of the matrix metallopeptidase 1 (MMP1) gene (G-[519]–C-[340]; OR 1.94, 95% CI 1.15–3.28; P = 0.013), while other haplotypes (A-[519]–C-[340] and G-[519]–T-[340], OR 0.68, 95% CI 0.52–0.88, P = 0.004, and OR 0.71, 95% CI 0.56–0.89, P = 0.003, respectively) provided protection when tested in two separate populations.⁴⁵ The entire family of MMP proteins are involved in the breakdown of extracellular matrix proteins, particularly collagen types I, II and III; thus MMP protein sequence changes can predispose individuals to plaque rupture.⁴⁵ Two SNPs in the cystathionine-beta-synthase (CBS) gene have been associated with decreased risk of coronary artery disease and increased responsiveness to the homocysteine-lowering effects of folic acid (699C>T and 1080T>C).⁴⁶ Interestingly, both SNPs are within the coding region but do not cause amino acid substitutions and do not affect the resulting protein, so are likely to be linked to other SNPs that affect the level or activity of CBS expression. In addition, a tag SNP representing a haplotype in the vitamin K epoxide reductase complex subunit 1 gene (VKORC1), which encodes the target for oral anticoagulants such as warfarin, has been associated with increased risk for stroke, vascular disease and aortic dissection.⁴⁷

Genetics is part of our clinical practice, but we have yet to exploit this field to improve our patient care. In the future, it will no longer be enough to obtain family history; we will carefully record the clinical phenotype and undertake further testing based on genetic findings.⁴⁸ We must be diligent in recording not only the diagnosis of hypertension or hyperlipidemia, but also what medications patients have received, whether treatment was successful, and to what degree. It is also important to document which side effects they have experienced, to be thorough when recording the findings of their cardiac catheterizations and echocardiograms, and to obtain as much information about their family tree as possible. A standard battery of future clinical tests might include testing for a group of hypertension haplotypes to help decide which medicines would be most beneficial; a group of high-risk coronary artery disease haplotypes to indicate if early, aggressive primary prevention is required; or a group of haplotypes that identify patients at risk for subacute thrombosis after stent implantation.

An example of an ongoing cardiovascular study in which the HapMap has a major role is the CardioGene study,⁴⁹ the primary goal of which is to identify genetic markers for in-stent restenosis in bare-metal stent implantation. This study will provide insight into how genetic markers can guide therapies in the catheterization laboratory and could become a model for how future projects can be performed on a larger scale. The CardioGene study was implemented in the fall of 2001 with prospective collection of DNA, RNA and serum in concert with extensive clinical phenotyping in almost 500 patients. Genotyping using the GeneChip^® Human Mapping 100K Set (Affymetrix Inc., Santa Clara, CA) was performed to identify predictive SNPs in patients with a susceptibility to in-stent restenosis. Individual SNPs have been analyzed in the context of the HapMap to identify susceptibility haplotypes with stronger predictive value than individual SNPs.⁵⁰ Confirmation of these findings will be vital and should be performed in well-established clinical registries.^{34, 51, 52} Comprehensive studies such as the CardioGene study are costly, labor-intensive and logistically difficult but could identify key genetic markers that will be extremely useful for clinical practice, and would not be possible without the HapMap. A key aspect of such studies is the collaboration between clinicians and basic scientists that will help bring the genomic revolution to patients. Scientific landmarks such as the HapMap are a further step towards bringing the bench to the bedside and will help propel research findings to the clinic.

Conclusion

The importance of the HapMap to cardiovascular research is increasingly clear. Genetic variants could explain why two patients with similar risk factors can have extremely dissimilar disease patterns, why two patients with a similar lipid profile can respond differently to statin therapy, why two patients presenting with acute myocardial infarction can have vastly different risk factor profiles and why two patients with similar clinical patterns can receive the same stent and have vastly different long-term results. Aspirin and clopidogrel resistance studies are driving the development of clinically useful bedside diagnostic testing, which will change the way we use these medications to treat our patients by titrating dose to response.⁵³ Registries and databases such as the National Heart, Lung and Blood Institute Dynamic Registry and the Framingham cohort, which have linked clinical data and DNA, are vital to discovery in cardiovascular genomics.^{3, 34, 51, 52} These databases could be poised to gain unique benefit from the HapMap, which will add a more efficient, less costly genetic component to the studies. Continued elegant phenotyping of patients, whether in the clinical notes in a patient's medical record or carefully recorded in a case report, is vital for genomic studies, with the goal of identifying the specific benefit for each patient.

Technology is moving forward quickly. Currently a DNA chip capable of identifying 500,000 genetic markers is available, which will increase the ability for SNP testing fivefold, and further advancements are on the horizon. We are only beginning to harness the power of genomics in cardiovascular disease following the discoveries in areas such as B-cell lymphoma.⁵⁴ Tools such as the HapMap are suited to decoding the complex, ubiquitous phenotypes, and once initial breakthroughs have been made we could begin to decode the complex disorders common in cardiovascular disease. How will the HapMap and the genomic revolution benefit our clinical practice? How will it change the way we prescribe medications and choose devices? The HapMap and the subsequent genomic advances in its wake will help us carry out what we do every day, only better.

References

1. Altshuler D et al. (2005) A haplotype map of the human genome. Nature 437: 1299–1320 | Article | PubMed | ISI | ChemPort |
2. Schmidt CW (2003) HapMap: building a database with blocks. EHP Toxicogenomics 111: A16 | PubMed |
3. The International HapMap Consortium (2003) The International HapMap Project. Nature 426: 789–796 | Article | PubMed | ISI | ChemPort |
4. Thorisson GA et al. (2005) The International HapMap Project web site. Genome Res 15: 1592–1593 | Article | PubMed | ChemPort |
5. The International HapMap Consortium (2004) Integrating ethics and science in the International HapMap Project. Nat Rev Genet 5: 467–475 | Article | ISI |
6. Kastrati A et al. (2006) Predictive factors of restenosis after coronary implantation of sirolimus- or paclitaxel-eluting stents. Circulation 113: 2293–2300 | Article | PubMed |
7. Di Castelnuovo A et al. (2005) Platelet glycoprotein IIb/IIIa polymorphism and coronary artery disease: implications for clinical practice. Am J Pharmacogenomics 5: 93–99 | Article | PubMed | ChemPort |
8. Koch W et al. (2004) Apolipoprotein E gene polymorphisms and thrombosis and restenosis after coronary artery stenting. J Lipid Res 45: 2221–2226 | PubMed | ChemPort |
9. Koch W et al. (2004) Peroxisome proliferator-activated receptor gamma gene polymorphisms and restenosis in diabetic patients after stenting in coronary arteries. Diabetologia 47: 1126–1127 | Article | PubMed | ChemPort |
10. Angiolillo DJ et al. (2004) 807 C/T Polymorphism of the glycoprotein Ia gene and pharmacogenetic modulation of platelet response to dual antiplatelet treatment. Blood Coagul Fibrinolysis 15: 427–433 | Article | PubMed | ChemPort |
11. Angiolillo DJ et al. (2004) High clopidogrel loading dose during coronary stenting: effects on drug response and interindividual variability. Eur Heart J 25: 1903–1910 | Article | PubMed | ChemPort |
12. Angiolillo DJ et al. (2004) PlA polymorphism and platelet reactivity following clopidogrel loading dose in patients undergoing coronary stent implantation. Blood Coagul Fibrinolysis 15: 89–93 | Article | PubMed |
13. Angiolillo DJ et al. (2004) Platelet aggregation according to body mass index in patients undergoing coronary stenting: should clopidogrel loading-dose be weight adjusted? J Invasive Cardiol 16: 169–174 | PubMed |
14. Chen SN et al. (2005) A common PCSK9 haplotype, encompassing the E670G coding single nucleotide polymorphism, is a novel genetic marker for plasma low-density lipoprotein cholesterol levels and severity of coronary atherosclerosis. J Am Coll Cardiol 45: 1611–1619 | Article | PubMed | ISI | ChemPort |
15. Bezzina CR et al. (2006) Common sodium channel promoter haplotype in Asian subjects underlies variability in cardiac conduction. Circulation 113: 338–344 | Article | PubMed | ChemPort |
16. Anderson JL et al. (2000) A common variant of the AMPD1 gene predicts improved cardiovascular survival in patients with coronary artery disease. J Am Coll Cardiol 36: 1248–1252 | Article | PubMed | ISI | ChemPort |
17. Hingorani AD et al. (1999) A common variant of the endothelial nitric oxide synthase (Glu298Asp) is a major risk factor for coronary artery disease in the UK. Circulation 100: 1515–1520 | PubMed | ISI | ChemPort |
18. Doris PA (2002) Hypertension genetics, single nucleotide polymorphisms, and the common disease:common variant hypothesis. Hypertension 39: 323–331 | Article | PubMed | ChemPort |
19. Samuels ME et al. (2001) Identification of a common variant in the lipoprotein lipase gene in a large Utah kindred ascertained for coronary heart disease: the -93G/D9N variant predisposes to low HDL-C/high triglycerides. Clin Genet 59: 88–98 | Article | PubMed | ChemPort |
20. Pritchard JK and Cox NJ (2002) The allelic architecture of human disease genes: common disease-common variant...or not? Hum Mol Genet 11: 2417–2423 | Article | PubMed | ISI | ChemPort |
21. Lunetta KL et al. (2004) Screening large-scale association study data: exploiting interactions using random forests. BMC Genet 5: 32 | Article | PubMed | ChemPort |
22. de Bakker PI et al. (2005) Efficiency and power in genetic association studies. Nat Genet 37: 1217–1223 | Article | PubMed | ISI | ChemPort |
23. Lander ES et al. (2001) Initial sequencing and analysis of the human genome. Nature 409: 860–921 | Article | PubMed | ISI | ChemPort |
24. ENCODE Project Consortium (2004) The ENCODE (ENCyclopedia Of DNA Elements) Project. Science 306: 636–640 | Article | PubMed | ISI | ChemPort |
25. Waterston RH et al. (2002) Initial sequencing and comparative analysis of the mouse genome. Nature 420: 520–562 | Article | PubMed | ISI | ChemPort |
26. Stoneking M (2001) Single nucleotide polymorphisms: from the evolutionary past. Nature 409: 821–822 | Article | PubMed | ISI | ChemPort |
27. The International HapMap Project [http://www.hapmap.org]
28. Agema WR et al. (2002) A meta-analysis of the angiotensin-converting enzyme gene polymorphism and restenosis after percutaneous transluminal coronary revascularization: evidence for publication bias. Am Heart J 144: 760–768 | Article | PubMed | ChemPort |
29. Amant C et al. (1997) D allele of the angiotensin I-converting enzyme is a major risk factor for restenosis after coronary stenting. Circulation 96: 56–60 | PubMed | ChemPort |
30. Rakugi H et al. (1994) Induction of angiotensin converting enzyme in the neointima after vascular injury. Possible role in restenosis. J Clin Invest 93: 339–346 | PubMed | ISI | ChemPort |
31. Wijpkema JS et al. (2006) Restenosis after percutaneous coronary intervention is associated with the angiotensin-II type-1 receptor 1166A/C polymorphism but not with polymorphisms of angiotensin-converting enzyme, angiotensin-II receptor, angiotensinogen or heme oxygenase-1. Pharmacogenet Genomics 16: 331–337 | PubMed | ChemPort |
32. Koch W et al. (2006) Association of transforming growth factor-1 gene polymorphisms with myocardial infarction in patients with angiographically proven coronary heart disease. Arterioscler Thromb Vasc Biol 26: 1114–1119 | Article | PubMed | ChemPort |
33. Koch WJ (2004) Genetic and phenotypic targeting of beta-adrenergic signaling in heart failure. Mol Cell Biochem 263: 5–9 | Article | PubMed | ChemPort |
34. Faxon DP et al. (1984) Determinants of successful percutaneous transluminal coronary angioplasty: report from the National Heart, Lung, and Blood Institute Registry. Am Heart J 108: 1019–1023 | Article | PubMed | ChemPort |
35. Holmes DR Jr et al. (1984) Restenosis after percutaneous transluminal coronary angioplasty (PTCA): a report from the PTCA Registry of the National Heart, Lung, and Blood Institute. Am J Cardiol 53: 77C–81C | Article | PubMed |
36. Venter JC et al. (2001) The sequence of the human genome. Science 291: 1304–1351 | Article | PubMed | ISI | ChemPort |
37. Goldstein DB and Weale ME (2001) Population genomics: linkage disequilibrium holds the key. Curr Biol 11: R576–R579 | Article | PubMed | ISI | ChemPort |
38. Smith AV et al. (2005) Sequence features in regions of weak and strong linkage disequilibrium. Genome Res 15: 1519–1534 | Article | PubMed | ISI | ChemPort |
39. Hinds DA et al. (2005) Whole-genome patterns of common DNA variation in three human populations. Science 307: 1072–1079 | Article | PubMed | ISI | ChemPort |
40. Nielsen R et al. (2005) Genomic scans for selective sweeps using SNP data. Genome Res 15: 1566–1575 | Article | PubMed | ChemPort |
41. Weir BS et al. (2005) Measures of human population structure show heterogeneity among genomic regions. Genome Res 15: 1468–1476 | Article | PubMed | ChemPort |
42. Forouhi NG and Sattar N (2006) CVD risk factors and ethnicity—a homogeneous relationship? Atheroscler 7 (Suppl): 11–19 | Article |
43. Bertoni AG et al. (2006) Diabetic cardiomyopathy and subclinical cardiovascular disease: the Multi-Ethnic Study of Atherosclerosis (MESA). Diabetes Care 29: 588–594 | Article | PubMed |
44. McCluskey J and Peh CA (1999) The human leucocyte antigens and clinical medicine: an overview. Rev Immunogenet 1: 3–20 | PubMed | ChemPort |
45. Pearce E et al. (2005) Haplotype effect of the matrix metalloproteinase-1 gene on risk of myocardial infarction. Circ Res 97: 1070–1076 | Article | PubMed | ChemPort |
46. Kruger WD et al. (2000) Polymorphisms in the CBS gene associated with decreased risk of coronary artery disease and increased responsiveness to total homocysteine lowering by folic acid. Mol Genet Metab 70: 53–60 | Article | PubMed | ChemPort |
47. Wang Y et al. (2006) VKORC1 haplotypes are associated with arterial vascular diseases (stroke, coronary heart disease, and aortic dissection). Circulation 113: 1615–1621 | Article | PubMed |
48. Weinshilboum R (2003) Inheritance and drug response. N Engl J Med 348: 529–537 | Article | PubMed | ISI |
49. Ganesh SK et al. (2004) Rationale and study design of the CardioGene Study: genomics of in-stent restenosis. Pharmacogenomics 5: 952–1004 | PubMed |
50. Ganesh SK and Skelding KA et al., Genome-wide association study of in-stent restenosis identifies multiple candidate susceptibility loci. Presented at the AHA Scientific Sessions: 2006 November 12–15, Chicago
51. Faxon DP et al. (1984) Role of percutaneous transluminal coronary angioplasty in the treatment of unstable angina. Report from the National Heart, Lung, and Blood Institute Percutaneous Transluminal Coronary Angioplasty and Coronary Artery Surgery Study Registries. Am J Cardiol 53: 131C–135C | Article | PubMed | ChemPort |
52. Lievers KJ et al. (2006) Association of a 31 bp VNTR in the CBS gene with postload homocysteine concentrations in the Framingham Offspring Study. Eur J Hum Genet 14: 1125–1129 | Article | PubMed | ChemPort |
53. Wang TH et al. (2005) Aspirin and clopidogrel resistance: an emerging clinical entity. Eur Heart J 27: 647–654 | PubMed |
54. Shivakumar L and Armitage JO (2006) Bcl-2 gene expression as a predictor of outcome in diffuse large B-cell lymphoma. Clin Lymphoma Myeloma 6: 455–457 | PubMed | ChemPort |

Contact the journal about this article

Subject areas under which this article appears: Genetics | Pathology