Susceptibility Loci For Complex Diseases Research Articles

Addressing the Missing Heritability Problem With the Help of Regulatory Features.

Genome-wide association studies (GWASs) have successfully identified thousands of susceptibility loci for human complex diseases. However, missing heritability is still a challenging problem. Considering most GWAS loci are located in regulatory elements, we recently developed a pipeline named functional disease-associated single-nucleotide polymorphisms (SNPs) prediction (FDSP), to predict novel susceptibility loci for complex diseases based on the interpretation of regulatory features and published GWAS results with machine learning. When applied to type 2 diabetes and hypertension, the predicted susceptibility loci by FDSP were proved to be capable of explaining additional heritability. In addition, potential target genes of the predicted positive SNPs were significantly enriched in disease-related pathways. Our results suggested that taking regulatory features into consideration might be a useful way to address the missing heritability problem. We hope FDSP could offer help for the identification of novel susceptibility loci for complex diseases.

Integrating regulatory features data for prediction of functional disease-associated SNPs.

Genome-wide association studies (GWASs) are an effective strategy to identify susceptibility loci for human complex diseases. However, missing heritability is still a big problem. Most GWASs single-nucleotide polymorphisms (SNPs) are located in noncoding regions, which has been considered to be the unexplored territory of the genome. Recently, data from the Encyclopedia of DNA Elements (ENCODE) and Roadmap Epigenomics projects have shown that many GWASs SNPs in the noncoding regions fall within regulatory elements. In this study, we developed a pipeline named functional disease-associated SNPs prediction (FDSP), to identify novel susceptibility loci for complex diseases based on the interpretation of the functional features for known disease-associated variants with machine learning. We applied our pipeline to predict novel susceptibility SNPs for type 2 diabetes (T2D) and hypertension. The predicted SNPs could explain heritability beyond that explained by GWAS-associated SNPs. Functional annotation by expression quantitative trait loci analyses showed that the target genes of the predicted SNPs were significantly enriched in T2D or hypertension-related pathways in multiple tissues. Our results suggest that combining GWASs and regulatory features data could identify additional functional susceptibility SNPs for complex diseases. We hope FDSP could help to identify novel susceptibility loci for complex diseases and solve the missing heritability problem.

Mapping eQTLs with RNA-seq reveals novel susceptibility genes, non-coding RNAs and alternative-splicing events in systemic lupus erythematosus.

Studies attempting to functionally interpret complex-disease susceptibility loci by GWAS and eQTL integration have predominantly employed microarrays to quantify gene-expression. RNA-Seq has the potential to discover a more comprehensive set of eQTLs and illuminate the underlying molecular consequence. We examine the functional outcome of 39 variants associated with Systemic Lupus Erythematosus (SLE) through the integration of GWAS and eQTL data from the TwinsUK microarray and RNA-Seq cohort in lymphoblastoid cell lines. We use conditional analysis and a Bayesian colocalisation method to provide evidence of a shared causal-variant, then compare the ability of each quantification type to detect disease relevant eQTLs and eGenes. We discovered the greatest frequency of candidate-causal eQTLs using exon-level RNA-Seq, and identified novel SLE susceptibility genes (e.g. NADSYN1 and TCF7) that were concealed using microarrays, including four non-coding RNAs. Many of these eQTLs were found to influence the expression of several genes, supporting the notion that risk haplotypes may harbour multiple functional effects. Novel SLE associated splicing events were identified in the T-reg restricted transcription factor, IKZF2, and other candidate genes (e.g. WDFY4) through asQTL mapping using the Geuvadis cohort. We have significantly increased our understanding of the genetic control of gene-expression in SLE by maximising the leverage of RNA-Seq and performing integrative GWAS-eQTL analysis against gene, exon, and splice-junction quantifications. We conclude that to better understand the true functional consequence of regulatory variants, quantification by RNA-Seq should be performed at the exon-level as a minimum, and run in parallel with gene and splice-junction level quantification.

Fine mapping of complex disease susceptibility loci

Genome-wide association studies (GWAS) using single nucleotide polymorphism (SNP) markers have identified more than 3800 susceptibility loci for more than 660 diseases or traits. However, the most significantly associated variants or causative variants in these loci and their biological functions have remained to be clarified. These causative variants can help to elucidate the pathogenesis and discover new biomarkers of complex diseases. One of the main goals in the post-GWAS era is to identify the causative variants and susceptibility genes, and clarify their functional aspects by fine mapping. For common variants, imputation or re-sequencing based strategies were implemented to increase the number of analyzed variants and help to identify the most significantly associated variants. In addition, functional element, expression quantitative trait locus (eQTL) and haplotype analyses were performed to identify functional common variants and susceptibility genes. For rare variants, fine mapping was carried out by re-sequencing, rare haplotype analysis, family-based analysis, burden test, etc.This review summarizes the strategies and problems for fine mapping.

Finding common susceptibility variants for complex disease: past, present and future

The identification of complex disease susceptibility loci has been accelerated considerably by advances in high-throughput genotyping technologies, improved insight into correlation patterns of common variants and the availability of large-scale sample sets. Linkage scans and small-scale candidate gene studies have now given way to genome-wide association scans. In this review, we summarize insights gained from the past, highlight practical issues relating to the design and analysis of current state-of-the-art GWA studies and look into future trends in the field of human complex trait genetics.

Using gene expression to investigate the genetic basis of complex disorders

The identification of complex disease susceptibility loci through genome-wide association studies (GWAS) has recently become possible and is now a method of choice for investigating the genetic basis of complex traits. The number of results from such studies is constantly increasing but the challenge lying forward is to identify the biological context in which these statistically significant candidate variants act. Regulatory variation plays an important role in shaping phenotypic differences among individuals and thus is very likely to also influence disease susceptibility. As such, integrating gene expression data and other disease relevant intermediate phenotypes with GWAS results could potentially help prioritize fine-mapping efforts and provide a shortcut to disease biology. Combining these different levels of information in a meaningful way is however not trivial. In the present review, we outline the several approaches that have been explored so far in this sense and their achievements. We also discuss the limitations of the methods and how upcoming technological developments could help circumvent these limitations. Overall, such efforts will be very helpful in understanding initially regulatory effects on disease and disease etiology in general.

The impact of data quality on the identification of complex disease genes: experience from the Family Blood Pressure Program

The application of genome-wide linkage scans to uncover susceptibility loci for complex diseases offers great promise for the risk assessment, treatment, and understanding of these diseases. However, for most published studies, linkage signals are typically modest and vary considerably from one study to another. The multicenter Family Blood Pressure Program has analyzed genome-wide linkage scans of over 12 000 individuals. Based on this experience, we developed a protocol for large linkage studies that reduces two sources of data error: pedigree structure and marker genotyping errors. We then used the linkage signals, before and after data cleaning, to illustrate the impact of missing and erroneous data. A comprehensive error-checking protocol is an important part of complex disease linkage studies and enhances gene mapping. The lack of significant and reproducible linkage findings across studies is, in part, due to data quality.

Identification of susceptibility loci for complex diseases in a case-control association study using the Genetic Analysis Workshop 14 dataset

Although current methods in genetic epidemiology have been extremely successful in identifying genetic loci responsible for Mendelian traits, most common diseases do not follow simple Mendelian modes of inheritance. It is important to consider how our current methodologies function in the realm of complex diseases. The aim of this study was to determine the ability of conventional association methods to fine map a locus of interest. Six study populations were selected from 10 replicates (New York) from the Genetic Analysis Workshop 14 simulated dataset and analyzed for association between the disease trait and locus D2. Genotypes from 45 single-nucleotide polymorphisms in the telomeric region of chromosome 3 were analyzed by Pearson's chi-square tests for independence to test for association with the disease trait of interest. A significant association was detected within the region; however, it was found 3 cM from the documented location of the D2 disease locus. This result was most likely due to the method used for data simulation. In general, this study showed that conventional case-control association methods could detect disease loci responsible for the development of complex traits.

Lee Responds to "Making the Most of Genotype Asymmetries"

I am grateful to Dr. Weinberg for her insightful commentary (1) on my paper (2). Weinberg is looking forward eagerly to a day when researchers will be able to search the whole genome for susceptibility loci for complex diseases (1). Likewise, I notice with excitement that the conventional “risk factor” epidemiology as we know it has undergone a profound change. Moving into this postgenomic era, epidemiology can mean genemapping business, no less truly than it has been for so long about odds ratios and confidence intervals for, for example, smoking and lung cancer. To obtain a weighting scheme for the various family configurations encountered in actual practice, the Di is defined as the allele count for the case (Ci) minus the allele count for the following: 1) nontransmitted, Ni (case-parents data); 2) imputed nontransmitted, (casesibling data); 3) spouse, Si (case-spouse data); and 4) imputed spouse, (case-offspring data), where the “nontransmitted” refers to the parental alleles not transmitted to the case (1), the “imputed nontransmitted” has an allele count of , where the Bi is the mean allele count of the control siblings, and the spouse and the imputed spouse are defined the same as in my paper (2). The Dis defined in this way all have the same “mean” (e1 in my paper (2)) under the alternative hypothesis, irrespective of family configurations. The statistical efficiencies of the various family configurations are then in proportion to the inverses of their respective “variances” (e2 in my paper (2)). Table 1 presents the relative efficiencies for the various family configurations (relative to case-parents data). Note that, in this paper, the disequilibrium test (1), but not the transmission/disequilibrium test (2), was applied to the case-parents data. To be brief, I present only the case with the allele frequency set at P = 0.1. It is of interest to note the following. First, the case-spouse data and the case-parents data have exactly the same efficiency. Second, the case-offspring data and the casesibling data have roughly the same relative efficiency, when the number per family of offspring and the number per family of control siblings are equal. Third, the relative efficiency of the case-offspring (case-sibling) data with one offspring (control sibling) per family is ~0.5. These findings largely confirm Weinberg’s speculations (1). In addition, I agree with Weinberg (1) that the information from the following two lines of relatives can be integrated to improve the power of the disequilibrium test: line I, the parents or, if the parents are missing, the control siblings; and line II, the spouse or, if the spouse is missing, the offspring. The penultimate column of table 1 presents the relative efficiencies (also calculated from e2 in my paper (2)) for a Di, averaged from caseparents and case-spouse data, and the last column, the relative efficiencies from one control sibling and one offspring. It is of interest to note that the relative efficiencies are less than 2.00 for parents plus spouse. (Having two controls does not make the case twice as informative). Moreover, the relative efficiencies are less than 1.00 for one control sibling plus one offspring. (Two haploids do not make a whole diploid control). The values are larger than the corresponding values for two control siblings or two offspring, though. (Having two lines of relatives is better than having only one). The variance formula simplifies considerably under the null. We have, dropping the subscript i, Var(D) = Var(C – N) = Var(C) + Var(N) = 4P(1 – P), for caseparents data. Likewise, Var(D) = Var(C – S) = Var(C) + Var(S) = 4P(1 – P), for case-spouse data. Because the numbers of “identical by descent” (3) for a sibling pair are 2 (probability = 0.25), 1 (probability = 0.5), and 0 (probability = 0.25), we have Cov(B1, B2) = Cov(C, B1) = 0.25 × 2P(1 – P) + 0.5 × P(1 – P) + 0.25 × 0 = P(1 – P), with B1 and B2 being the allele count for the first and the second control siblings, respectively. Thus, for casesibling data with x control siblings per family,

It's Not Just the Genes

T he most common diseases are the toughest to crack. Heart disease, cancer, diabetes, psychiatric illness: All of these are “complex” or “multifactorial” diseases, meaning that they cannot be ascribed to mutations in a single gene or to a single environmental factor. Rather they arise from the combined action of many genes, environmental factors, and risk-conferring behaviors. One of the greatest challenges facing biomedical researchers today is to sort out how these contributing factors interact in a way that translates into effective strategies for disease diagnosis, prevention, and therapy. The genes that contribute to complex disease are notoriously difficult to identify, because they typically exert small effects on disease risk; in addition, the magnitude of their effects is likely to be modified by other unrelated genes as well as environmental factors. Perhaps reflecting these difficulties, susceptibility loci for complex diseases identified in one study population often cannot be replicated in other populations (see the Report by Levinson et al. , p. [739][1]). This special section describes three complex diseases that together illustrate the difficulty facing researchers. Type II diabetes (see the News story by Marx, p. [686][2]) has reached epidemic proportions in the United States and, in a particularly disturbing trend, is now striking people at a young age. The clear physiological link between this form of diabetes and obesity, whose incidence is also on the rise, makes a strong case for a causal role of diet and exercise level; at the same time, studies of diabetes-prone human populations and rodent models have begun to uncover tantalizing candidate genes that are also likely to influence disease susceptibility. In patients suffering from the autoimmune disorder lupus (see the News story by Marshall, p. [689][3]), the immune system recognizes host tissue as foreign, sometimes producing serious organ damage. The working model for disease etiology suggests that people with lupus are genetically predisposed to have a hyperactive immune system, but the manifestation of symptoms is also influenced by environmental factors, including viruses, drugs, and exposure to sunlight. Sawa and Snyder (p. [692][4]) review our current understanding of schizophrenia, a debilitating psychiatric disorder affecting an estimated 1% of the population. Intriguingly, three diverse investigative approaches—pharmacological analysis, brain imaging, and genetic studies—are implicating the same neurotransmitter systems in the disease, although the role of environmental factors such as viruses and birth trauma has not been excluded. In a series of commentaries, three scientists who do not subscribe to the current “genocentric” view of disease argue that progress in understanding complex diseases will depend on additional branches of science. Willett (p. [695][5]) discusses the importance of carefully designed epidemiological studies in disease-prevention research. Such studies have revealed that more than 70% of stroke, colon cancer, coronary heart disease, and type II diabetes is potentially preventable by life-style modifications. Rees (p. [698][6]) discusses the critical role physicians play in diagnosing and treating patients with complex diseases. He notes that skilled clinicians have designed very effective therapies for common skin disorders such as psoriasis and acne simply through careful observational studies of patients and without any knowledge of the contributory genes. Finally, Strohman (p. [701][7]) points out that systems-based approaches such as metabolic control analysis can yield important insights into complex diseases; however, progress in this arena will depend on the development of new technologies and better training of biomedical scientists in mathematics and quantitative biochemistry. Can the puzzle of complex diseases be solved? With integrated approaches and coordinated efforts from researchers in diverse disciplines, there is much room for optimism. [1]: /lookup/doi/10.1126/science.1069914 [2]: /lookup/doi/10.1126/science.296.5568.686 [3]: /lookup/doi/10.1126/science.296.5568.689 [4]: /lookup/doi/10.1126/science.1070532 [5]: /lookup/doi/10.1126/science.1071055 [6]: /lookup/doi/10.1126/science.296.5568.698 [7]: /lookup/doi/10.1126/science.1070534

Clustering of pedigrees using marker allele frequencies: impact on linkage analysis.

Ethnicity may form the basis for locus heterogeneity at certain susceptibility loci for complex diseases. Classification of pedigrees into ethnic groups is usually based upon self-report, but this may not be sensitive or specific. We investigated whether it is possible to cluster families from an admixed population using pedigree-specific marker allele frequencies. We used 323 autosomal microsatellite markers from 216 pedigrees who described themselves as either Caucasian or African American. First, we compared the stated ethnicity of pedigrees with clusters using pedigree-specific marker allele frequencies as input for a self-organizing map, a type of neural network. Using data from different chromosomes, nine pedigrees which were self-reported as African American were clustered with the Caucasian pedigrees. Removal of these nine pedigrees from the African American group did not markedly affect linkage results. We then proceeded to determine whether there was further heterogeneity between pedigrees using 1 x 3 nodes. Forty-four pedigrees were clustered in a group intermediate to the African American or Caucasian clusters. This group was composed of 36 and 8 pedigrees that described themselves as African American and Caucasian, respectively. Linkage analysis was performed in this group and results compared with the groups based upon self-reported ethnicity. Linkage to a region on chromosome 3 was observed in this intermediate group, which was more significant than any of the results obtained when pedigrees were grouped using self-reported ethnicity. Use of marker data may assist in clustering pedigrees with similar, ethnic backgrounds and may increase the power for genetic linkage studies.

6 Definition of the phenotype

Definition of the phenotype is a key issue in designing any genetic study whose goal is to detect disease genes. This chapter describes strategies to increase the power to detect susceptibility loci for complex diseases. A narrowly defined disease phenotype can offer advantages over broad definitions. Studies of clinical disease can also benefit from judicious selection of endophenotypes and related quantitative traits for analysis. The effect of diagnostic and measurement error is also discussed; power is maximized when strategies to reduce error are incorporated into a study design.