Shaking the tree: mapping complex disease genes with linkage disequilibrium

A gene-centric approach to genome-wide association studies

Genic variants are more likely to alter gene function and affect disease risk than those that occur outside genes. Variants in genes, however, might not be sufficiently covered by the existing approaches to genome-wide association studies. Our analysis of the HapMap ENCODE data indicates that this concern is valid, and that an alternative approach that focuses on genic variants provides a more complete coverage of functionally important regions and a greater genotyping efficiency. We therefore argue that resources should be developed to make gene-centric genome-wide association studies feasible.

Genomics and complex liver disease: Challenges and opportunities

The concept of genetic susceptibility in the contribution to human disease is not new. What is new is the emerging ability of the field of genomics to detect, assess, and interpret genetic variation in the study of susceptibility to development of disease. Deciphering the human genome sequence and the publication of the human haplotype map are key elements of this effort. However, we are only beginning to understand the contribution of genetic predisposition to complex liver disease through its interaction with environmental risk factors. In the coming decade, we anticipate the development of human studies to better dissect the genotype/phenotype relationship of complex liver diseases. This endeavor will require large, well-phenotyped patient populations of each disease of interest and proper study designs aimed at answering important questions of hepatic disease prognosis, pathogenesis, and treatment. Teamwork between patients, physicians, and genomics scientists can ensure that this opportunity leads to important biological discoveries and improved treatment of complex disease.

Mapping trait loci by use of inferred ancestral recombination graphs

Large-scale association studies are being undertaken with the hope of uncovering the genetic determinants of complex disease. We describe a computationally efficient method for inferring genealogies from population genotype data and show how these genealogies can be used to fine map disease loci and interpret association signals. These genealogies take the form of the ancestral recombination graph (ARG). The ARG defines a genealogical tree for each locus, and, as one moves along the chromosome, the topologies of consecutive trees shift according to the impact of historical recombination events. There are two stages to our analysis. First, we infer plausible ARGs, using a heuristic algorithm, which can handle unphased and missing data and is fast enough to be applied to large-scale studies. Second, we test the genealogical tree at each locus for a clustering of the disease cases beneath a branch, suggesting that a causative mutation occurred on that branch. Since the true ARG is unknown, we average this analysis over an ensemble of inferred ARGs. We have characterized the performance of our method across a wide range of simulated disease models. Compared with simpler tests, our method gives increased accuracy in positioning untyped causative loci and can also be used to estimate the frequencies of untyped causative alleles. We have applied our method to Ueda et al.'s association study of CTLA4 and Graves disease, showing how it can be used to dissect the association signal, giving potentially interesting results of allelic heterogeneity and interaction. Similar approaches analyzing an ensemble of ARGs inferred using our method may be applicable to many other problems of inference from population genotype data.

Associations of plasma fibrinogen and factor VII clotting activity with coronary heart disease and stroke: prospective cohort study from the screening phase of the Thrombosis Prevention Trial

BACKGROUND

As with 'conventional' risk factors such as cholesterol and smoking, there is a need for large, long-term prospective studies on hemostatic factors.

OBJECTIVES

To investigate the prospective relationship of fibrinogen and factor VII clotting activity (FVIIc) with risk of coronary heart disease (CHD) and stroke in a study with a large number of outcomes over a period of 15 years.

PATIENTS/METHODS

A cohort of 22 715 men aged 45-69 years was screened for participation in the Thrombosis Prevention Trial. Men were followed up for fatal and non-fatal CHD and stroke events. There were 1515 CHD events (933 CHD deaths) and 391 strokes (180 stroke deaths). Hazard ratios (HRs) and 95% confidence intervals are expressed per standardized increase in log fibrinogen and log FVIIc, adjusting for age, trial treatment group, conventional CHD risk factors and regression dilution bias.

RESULTS

Hazard ratios for fibrinogen were 1.52 (1.37-1.70) for all CHD events, and 1.36 (1.09-1.69) for all strokes. Exclusion of events within the first 10 years showed a persistent association between CHD and fibrinogen, with an adjusted HR of 1.93 (1.42-2.64). The HRs for FVIIc, adjusting for age and trial treatment, were 1.07 (1.01-1.12) for all CHD events and 1.07 (0.97-1.20) for all strokes, and the fully adjusted HRs were, respectively, 0.97 (0.84-1.05) and 1.07 (0.85-1.33).

CONCLUSIONS

The persisting association between fibrinogen and CHD beyond 10 years may imply a causal effect. There is a small effect of FVIIc on CHD, after adjustment for age and trial treatment, but no association independent of other risk factors.

Implications of small effect sizes of individual genetic variants on the design and interpretation of genetic association studies of complex diseases

Accumulated evidence from searching for candidate gene-disease associations of complex diseases can offer some insights as the field moves toward discovery-oriented approaches with massive genome-wide testing. Meta-analyses of 50 non-human lymphocyte antigen gene-disease associations with documented overall statistical significance (752 studies) show summary odds ratios with a median of 1.43 (interquartile range, 1.28-1.65). Many different biases may operate in this field, for both single studies and meta-analyses, and these biases could invalidate some of these seemingly "validated" associations. Studies with a sample size of >500 show a median odds ratio of only 1.15. The median sample size required to detect the observed summary effects in each population addressed in the 752 studies is estimated to be 3,535 (interquartile range, 1,936-9,119 for cases and controls combined). These estimates are steeply inflated in the presence of modest bias. Population heterogeneity, as well as gene-gene and gene-environment interactions, could steeply increase these estimates and may be difficult to address even by very large biobanks and observational cohorts. The one visible solution is for a large number of teams to join forces on the same research platforms. These collaborative studies ideally should be designed up front to also assess more complex gene-gene and gene-environment interactions.

Genetic susceptibility to infectious diseases: big is beautiful, but will bigger be even better?

Genetic epidemiology, including twin studies, provides robust evidence that genetic variation in human populations contributes to susceptibility to infectious disease. One of the major limitations of studies that attempt to identify the genes and mechanisms that underlie this susceptibility has been lack of power caused by small sample size. With the development of novel technologies, burgeoning information on the human genome, the HapMap project, and human genetic diversity, we are at the beginning of a new era in the study of the genetics of complex diseases. This review looks afresh at the epidemiological evidence that supports a role for genetics in susceptibility to infectious disease, examines the somewhat limited achievements to date, and discusses current advances in methodology and technology that will potentially lead to translational data in the future.

Resources for Genetic Variation Studies

The rapid growth of genome-wide diversity databases, as well as ongoing large-scale resequencing projects targeting genes and other functional components of our genome, provide valuable resources of natural variation at the DNA sequence level. In this review, we briefly summarize the wealth of data on DNA polymorphisms in humans, the distribution of this diversity in the genome as well as among individuals, and the consequence of recombination on its organization. These data provide a set of powerful tools that can be used to better understand inherited phenotypic variation in humans. We discuss the implications for the design of studies investigating correlations between genotypes and phenotypes, both at the fundamental level of genome function and regulation, and for the mapping of disease genes.

A645G (Lys216Glu) polymorphism of the bactericidal/permeability-increasing protein gene in periodontal disease

Bactericidal/permeability-increasing protein (BPI) is a member of the pattern recognition receptors of the innate immune system and recognizes lipopolysaccharides (LPS), a bacterial component belonging to the pathogen-associated molecular patterns (PAMPs). BPI mediates the neutralization of LPS and increases the phagocytosis and cytotoxicity against bacteria. Recently, the functionally effective polymorphism A645G resulting in the amino acid alteration Lys216Glu has been described. The aim of the study was to investigate the association of the A645G polymorphism with chronic periodontal disease. The study population comprised 123 patients with periodontal disease (36 with mild, 52 with moderate and 35 with severe periodontitis) and 122 healthy, unrelated control individuals. Genotyping of the BPI gene polymorphism A645G (Lys216Glu) was performed by polymerase chain reaction and restriction fragment length polymorphism analysis. Statistical analysis was carried out employing the chi(2) test with Yates correction. Genotype and allele frequencies of the polymorphism tested herein showed no significant differences between periodontal disease as compared to the control group. The frequencies of the G allele were 52.4% in patients with periodontal disease and 49.2% in the control individuals (P = 0.528). Moreover, no significant associations could be detected after stratification for disease severity and according to gender. The present study does not give evidence for the contribution of the BPI gene to the genetic background of chronic periodontal disease.

Application of microarray technology in psychotropic drug trials

Microarrays can be manufactured to detect hundreds of thousands of polymorphisms in DNA from patients in psychotropic drug trials. Some of these polymorphisms may be useful as pharmacogenetic predictors of treatment outcomes. We tested a microarray designed to detect common polymorphisms in the CYP2D6 gene that encodes debrisoquine hydroxylase (DH). DH is involved in the hepatic metabolism of many psychotropics. CYP2D6 genotypes predicted plasma steady state concentrations of nortriptyline, a classic DH substrate, in a sample of geriatric patients with major depression. However, in a sample of 246 geriatric patients treated with paroxetine or mirtazapine, both of which are metabolized in part by DH, CYP2D6 genotypes determined with microarrays did not predict discontinuations due to adverse events or severity of adverse events. For modern antidepressants such as paroxetine and mirtazapine, pharmacokinetic factors that are regulated by CYP2D6 such as plasma drug concentrations may be less important than pharmacodynamic factors in determining outcomes. Studies of single candidate genes such as CYP2D6 have only begun to utilize the potential of microarrays for pharmacogenetic prediction. Yet, there is controversy as to whether genome-wide studies designed to detect millions of genotypes with microarrays will lead to new pharmacogenetic discoveries, or whether a more focused, hypothesis-driven approach is better.

Complex disease-associated pharmacogenetics: drug efficacy, drug safety, and confirmation of a pathogenetic hypothesis (Alzheimer's disease)

Safety and efficacy pharmacogenetics can be applied successfully to the drug discovery and development pipeline at multiple phases. We review drug-target screening using high throughput SNP associations with complex diseases testing more than 1,800 candidate targets with approximately 7,000 SNPs. Alzheimer's disease data are provided as an example. The supplementation of target-selected screening with genome-wide SNP association, to also define susceptibility genes and relevant disease pathways for human diseases, is discussed. Applications for determining predictive genetic or genomic profiles, or derived biomarkers, for drug efficacy and safety during clinical development are exemplified by several successful experiments at different phases of development. A Phase I-IIA study of side effects using an oral drug for the treatment of breast cancer is used as an example of early pipeline pharmacogenetics to predict side effects and allow optimization of dosing. References are provided for several other recently published genetic association studies of adverse events during drug development. We illustrate the early identification of gene variant candidates related to efficacy in a Phase IIA obesity drug trial to generate hypotheses for testing in subsequent development. How these genetic data generated in Phase IIA are subsequently incorporated as hypotheses into later Phase clinical protocols is discussed. A Phase IIB clinical trial for Alzheimer's disease is described that exemplifies the major pipeline decision between program attrition and further clinical development. In this case, there was no significant improvement in 511 intention-to-treat patients but, applying a confirmed prognostic biomarker (APOE4) to segment the clinical trial population, all three doses of rosiglitazone demonstrated improvement in patients who did not carry the APOE4 allele. The data for the APOE4 carriers demonstrated no significant improvement but suggested that there may be a need for higher doses. Thus, a development program that would have been terminated progressed to Phase III registration trials based on the results of prospective efficacy pharmacogenetic analyses. The implications of using APOE genotype as a biomarker to predict efficacy and possibly dose, as well as supporting the basic neurobiology and pharmacology that provided the original target validation, is discussed. Citations are provided that support a slow neurotoxic effect over many years of a specific fragment of apoE protein (over-produced by apoE4 substrate compared to apoE3) on mitochondria and the use of rosiglitazone to increase mitochondrial biogenesis and improve glucose utilization. Pharmacogenetics is currently being used across the pipeline to prevent attrition and to create safer and more effective medicines.

Genetic dissection of the common epilepsies

PURPOSE OF REVIEW

Only two functionally validated susceptibility genes, CACNA1H and GABRD, have so far been identified in the common epilepsies using a candidate gene approach. The difficulty with the alternative statistical approach, where none of the suggested candidates has been functionally validated, may partly be due to the posited genetic architecture of the common epilepsies, such as the idiopathic generalized epilepsies. A subset of both rare and common variants from a much larger pool of susceptibility genes may contribute to disease risk. We review methods and designs for the genetic dissection of common epilepsies.

RECENT FINDINGS

Genetic association studies, though theoretically more powerful than linkage analysis, have not yet delivered validated susceptibility genes. Methodological flaws can undermine such studies but are correctable. Concerns remain, however, about the extent of underlying genetic heterogeneity in common epilepsies. Genome-wide association studies are increasingly feasible, but issues remain about their conduct and analysis. Meta-analysis may resolve conflicting association studies, facilitated by the establishment of databases of genetic association studies. Newer multi-locus and admixture mapping approaches are attractive alternatives to traditional association studies and may offer new insights into identifying epilepsy genes.

SUMMARY

We conclude by emphasizing the importance of deeper endophenotyping using electroclinical, imaging, and molecular approaches to dissect the common epilepsies.

Evaluating coverage of genome-wide association studies

Genome-wide association studies involving hundreds of thousands of SNPs in thousands of cases and controls are now underway. The first of many analytical challenges in these studies involves the choice of SNPs to genotype. It is not practical to construct a different panel of tag SNPs for each study, so the first generation of genome-wide scans will use predefined, commercially available marker panels, which will in part dictate their success or failure. We compare different approaches in use today, and show that although many of them provide substantial coverage of common variation in non-African populations, the precise extent is strongly dependent on the frequencies of alleles of interest and on specific considerations of study design. Overall, despite substantial differences in genotyping technologies, marker selection strategies and number of markers assayed, the first-generation high-throughput platforms all offer similar levels of genome coverage.

Genetics of type 2 diabetes

After several years of uncertain progress, the stage is now set for a transformation in understanding the genetic landscape of type 2 diabetes. Advances in genome informatics, genotyping technology, and statistical methodology, allied to availability of large-scale clinical material, are having a salutary effect on susceptibility gene discovery. The advent of genuinely genome-wide association scans and the prospects for combining genetics with high-throughput genomics are additional sources of optimism for the future.

Genetic factors in the predisposition to drug-induced hypersensitivity reactions

Drug hypersensitivity reactions can occur with most drugs, although the frequency, severity, and clinical manifestations vary. Case reports have suggested that there may be familial clustering of drug hypersensitivity suggesting a genetic predisposition. As with most other forms of drug response, predisposition to drug hypersensitivity reactions is likely to be multifactorial and multigenic. Given the immune pathogenesis of these reactions, it is perhaps not surprising that the most significant genetic associations have been identified in the major histocompatibility complex for drugs such as abacavir, carbamazepine, and allopurinol. For abacavir, it has been suggested that preprescription genotyping for HLA-B*5701 in whites may reduce the incidence of hypersensitivity. It is likely that as our knowledge of variation in the human genome improves, coupled with improvements in technology, many more significant genetic predisposing factors for drug hypersensitivity are likely to be identified in the next decade. However, as we search for these genetic factors, it is important that we do not forget environmental predisposition, and to bear in mind that a genetic marker for drug hypersensitivity in one population may not necessarily be relevant for another population. Notwithstanding the advances in genetic technologies, the ultimate determinant of success in this area of research will be the identification and careful phenotyping of patients with drug hypersensitivity reactions. As we progress to whole genome scanning, in order to satisfy the requirements for adequate statistical power, the identification of large numbers of carefully phenotyped patients will be feasible only through international collaborations.

Genetic epidemiology and public health: hope, hype, and future prospects

Genetic epidemiology is a rapidly expanding research field, but the implications of findings from such studies for individual or population health are unclear. The use of molecular genetic screening currently has some legitimacy in certain monogenic conditions, but no established value with respect to common complex diseases. Personalised medical care based on molecular genetic testing is also as yet undeveloped for common diseases. Genetic epidemiology can contribute to establishing the causal nature of environmentally modifiable risk factors, through the application of mendelian randomisation approaches and thus contribute to appropriate preventive strategies. Technological and other advances will allow the potential of genetic epidemiology to be revealed over the next few years, and the establishment of large population-based resources for such studies (biobanks) should contribute to this endeavour.

Key concepts in genetic epidemiology

This article is the first in a series of seven that will provide an overview of central concepts and topical issues in modern genetic epidemiology. In this article, we provide an overall framework for investigating the role of familial factors, especially genetic determinants, in the causation of complex diseases such as diabetes. The discrete steps of the framework to be outlined integrate the biological science underlying modern genetics and the population science underpinning mainstream epidemiology. In keeping with the broad readership of The Lancet and the diverse background of today's genetic epidemiologists, we provide introductory sections to equip readers with basic concepts and vocabulary. We anticipate that, depending on their professional background and specialist knowledge, some readers will wish to skip some of this article.

Recent developments in genomewide association scans: a workshop summary and review

With the imminent availability of ultra-high-volume genotyping platforms (on the order of 100,000-1,000,000 genotypes per sample) at a manageable cost, there is growing interest in the possibility of conducting genomewide association studies for a variety of diseases but, so far, little consensus on methods to design and analyze them. In April 2005, an international group of >100 investigators convened at the University of Southern California over the course of 2 days to compare notes on planned or ongoing studies and to debate alternative technologies, study designs, and statistical methods. This report summarizes these discussions in the context of the relevant literature. A broad consensus emerged that the time was now ripe for launching such studies, and several common themes were identified--most notably the considerable efficiency gains of multistage sampling design, specifically those made by testing only a portion of the subjects with a high-density genomewide technology, followed by testing additional subjects and/or additional SNPs at regions identified by this initial scan.