It is widely believed that risks of many complex diseases are determined by genetic susceptibilities, including environmental exposures, and their interaction. Chatterjee and Carroll (2005) have recently developed an efficient retrospective maximum-likelihood method for analysis of case-control studies that exploits an assumption of gene-environment independence and leaves the distribution of the environmental covariates to be completely nonparametric. We generalize the semiparametric maximum-likelihood approach to situations when some of the environmental covariates are measured with error and allow genetic information to be missing on some subjects, e.g., unphased haplotypes. Profile likelihood techniques and an EM algorithm are developed, resulting in a relatively simple procedure for parameter estimation. We prove consistency and derive the resulting asymptotic covariance matrix of parameter estimates when variance of measurement error is known and when it is estimated using replications. The performance of the proposed method is illustrated using simulation studies emphasizing the case when genetic information is in the form of a haplotype and missing data arises from haplotype-phase ambiguity and missing genetic data. Inference is performed via a likelihood-ratio type procedure, one that we show has better small-sample performance thanWald-type inferences. An application of this method is illustrated using a case-control study of an association of calcium intake with early stages of colorectal tumor development.

Gene-environment interactions are an area of increasing interest in complex hu- man diseases. The first step in any study of the interactions between genes and the environment involves identifying genes which influence the trait of interest. In this dissertation, a new method for using the information in complex pedigrees to per- form a joint linkage disequilibrium and linkage mapping of quantitative trait loci is developed. Subsequently, methods are needed to determine the interaction, if any, between these genes and environmental risk factors. Many of these factors, such as weight or age, are continuous and little is known about their distributions. Thus, we introduce a new method for estimating the gene-environment interaction parameters in a logistic regression for the case-control study design. In doing so, we make the assumption that in the underlying population, the distributions of the genetic factors and the environmental covariates are independent. Additionally, we treat the envi- ronmental parameters nonparametricly, utilizing the profile likelihood. Furthermore, the methodology we develop is also general enough to be used on many different types of genetic information, including haplotypes, and can accommodate missing genotype data. The method is also extended to allow analysis in the presence of population stratification or genotype misclassification. We show that the standard errors of pa- rameter estimates using our method are smaller than those found using complete data only. These methods are illustrated using simulations and are applied to a real data set exploring the interaction between genotype and environment in disease risk.

Introduction: There is evidence to suggest that common genetic polymorphisms can modify the effect of environmental risk factors on colorectal neoplasia. Methods: Data on 1430 individuals were obtained from two chemoprevention trials, the Wheat Bran Fiber Trial (WBF) (1) and the Effects of Ursodeoxycholic Acid on Adenomatous Polyp Recurrence Trial (URSO) (2). Data were analyzed to test for gene-environment interactions between allelic variation in PPAR-γ (Pro12Ala, C1431T), body mass index (BMI) and waist circumference, and the biochemical biomarkers of metabolic syndrome. Simulated data were then used to determine if the sample size required to formally test the relationship between gene-exposure interactions could be reduced by using a genetically enriched study population. For this simulation aspect of the work, an established gene-drug interaction (i.e.: flavin monooxygenase 3 (FMO3) and sulindac) was used as a model system. Results: There was a borderline significant interaction between BMI and PPAR-γ for the Pro12Ala genotype (p(inter)=0.11) and significant interactions between BMI and the C1431T genotype (p(inter)=0.09). Results from the recursive partition model identified BMI (p = 0.007) and baseline fasting glucose levels (p =0.033) as significant predictors of colorectal adenoma recurrence for carriers of Ala12 and waist circumference as a significant predictor for the Pro12Pro12 carriers (p=0.002). Results from the simulated studies indicated that using genetically pre-screened and enriched populations can result in a 50% savings in the number of subjects required to test the candidate gene-drug interaction. Conclusions: These findings provide evidence that the effect of allelic variation in PPAR-γ on colorectal adenoma recurrence is modified by BMI and that component traits of metabolic syndrome differentially affect the risk of colorectal adenoma recurrence depending on genotype. In addition, using genotype as an inclusion criterion in future studies of adenoma recurrence will result in a smaller sample size required to test gene-environment interactions.

Estimations of heritability and variance explained due to environmental exposures and interaction effects help in understanding complex diseases. Current methods to detect such interactions rely on variance component methods. These methods have been neces- sary due to the m » n problem, where the number of predictors (m) vastly outnumbers the number of observations (n). These methods are all computationally intensive, which is further exacerbated when considering gene-environment interactions, as the number of predictors increases from m to 2m+1 in the case of a single environmental exposure. Novel methods are thus needed to enable fast and unbiased calculations of the variance explained (R2) for gene-environment interactions in very large samples on multiple traits. Taking advantage of the large number of participants in contemporary genetic studies, we herein propose a novel method for continuous trait R2 estimates that are up to 20 times faster than current methods. We have devised a novel method, monsterlm, that enables multiple linear regression on large regions encompassing tens of thousands of variants in hundreds of thousands of participants. We tested monsterlm with simulations using real genotypes from the UK Biobank. During simulations we verified the properties of monsterlm to estimate the variance explained by interaction terms. Our preliminary results showcase potential interactions between blood biochemistry biomarkers such as HbA1c, Triglycerides and ApoB with an environmental factor relating to obesity-related lifestyle factor: Waist-hip Ratio (WHR). We further investigate these results to reveal that more than 50% of the interaction variance calculated can be attributed to ∼5% of the single-nucleotide polymorphisms (SNPs) interacting with the environmental trait. Lastly, we showcase the impact of interactions on improving polygenic risk scores. / Thesis / Master of Science (MSc)

No / This article describes how the teaching of variation and genetics can give rise to the mistaken idea that genes are the sole determinants of our characteristics, that genes work in isolation to produce genetic disorders, such as cystic fibrosis. It goes on to discuss examples of gene environment interactions that give a more relevant and realistic account of how genes and environment interact in human genetic disease and stem cell technology. Finally, a conceptual model is introduced that might be useful for teaching, in which genes and environment are given equal status in explaining development.

This study examined possible interactions between DRD4 genotype and parenting on children’s externalizing, internalizing, and prosocial behaviors, and explored both parent and child perceptions of all variables. Both diathesis-stress and differential susceptibility hypotheses were assessed for examining possible interactions for children with the DRD4 7-repeat allele. Data were collected from 58 families within the Southern Illinois Twins/Triplets and Siblings Study (SITSS). Results indicated that although no gene-environment interactions were found when examining child perceptions, a significant interaction emerged between DRD4 and parenting in predicting externalizing behaviors when using parent reports. Children without the 7-repeat allele appeared to be malleable to both positive and negative parenting, supporting differential susceptibility. Also, child and parent reports of parenting were both predictive of child behavior. Lastly, MZ twins perceived more similar parenting environments than DZ twins, and there appeared to be a stronger environmental effect of parenting on externalizing after controlling for the effects of genes. This study adds to the differential susceptibility literature and points to the importance of considering perceptions of both children and parents when examining the effects of parenting on child behaviors.

No description available.

Genome-wide association studies in multifactorial diseases have contributed to our understanding of genetic risk and defined specific disease-associated loci in particular populations. However, risk cannot be fully explained by genetics and evidence points to both genetic and environmental factors being important in causation and pathophysiology. The role of vitamin D in calcium homeostasis is well established. Over the last 30 years it has become clear that vitamin D has a wider role in physiology and disease, notably in autoimmune, cancer and infectious conditions. However, the molecular mechanisms and possible causal role of these associations are poorly understood. Here I propose that the role of vitamin D in immune and inflammatory responses is significant, that genetic variation partly determines the response to vitamin D and that integrative analysis can yield important insights for disease mechanisms. For this I investigate the relationship between vitamin D and genetic risk involving the immune system by focusing on multiple sclerosis and sepsis, conditions classically defined as autoimmune and inflammatory respectively. I describe data resolving genetic variation associated with autoimmune diseases in vitamin D receptor binding sites; the association to multiple sclerosis of a genetic variant lying within a VDR binding site; the correlation of plasma vitamin D with genotype and cell specific transcriptomes in healthy volunteers; and the extent of vitamin D deficiency in severe sepsis and septic shock, its association with survival, correlation with gene expression and use in sub-classification to identify patients at higher risk of death. The limitations of each study and future work are discussed. Integrating epidemiological and clinical observations with genetic and functional genomics techniques has the potential to reveal interactions in population specific disease susceptibility that may lead to an improved understanding of disease mechanisms and clinical translation. The work I present here bridges molecular analysis, candidate and genome- wide, with phenotypic observations that are important in our understanding of disease.

Tumor-like proliferative vascular lesions manifest in several diseases such as peripheral arterial disease (PAD) and atherosclerosis (AS) after arterial injury. The cause of the vascular cell dysfunction in PAD patients is not known. Our recent novel discovery shows that inhibitor of differentiation 3 (ID3) is highly expressed in intimal lesions of clinical vascular disease samples. The central hypothesis of our study is: estrogenic chemical induced dysregulation of ID3 target genes is involved in the development of vascular disease. NHANES data analysis demonstrated higher geometric levels of all 6 PCB congeners in both PAD diagnosed participants and participants at risk of AS when compared to the rest of the population. Adjusted models showed association between higher exposure of PCBs, phthalates, BPA, and increased risk of PAD. Furthermore PCB153 was shown to have the highest geometric mean amongst all PCB congeners in both participants diagnosed with PAD and at risk of AS. Gene expression of ID3 & ID3 candidate targets in blood & tissue studies identified ID3 & ID3 candidate target genes as a driver of vascular disease. Overlapping ID3 & ID3 candidate target genes included: ABCB6, ACP1, BYSL, CAD, CDH15, DCBLD2, DHRS3, DNMT1, ID3, MCM4, and NDUFA7. The ID3 target genes involved in the: focal adhesion pathway were ACTN1, COL1A2, COL3A1, COL6A1, CTNNB1, IBSP, ID3, ITGA8, and MYL2; ECM-receptor interaction were COL1A2, COL3A1, COL6A1, IBSP, ID3, and ITGA8; oxidative phosphorylation pathway ATP5D, ATP5H, ATP6V0B, ATP6V0D1, ATP6V1B2, COX5A, COX7C, COX8A, CYC1, ID3, NDUFA1, NDUFA7, NDUFS4, NDUFV1, NDUFV2; and cell cycle pathway ANAPC10, ATM, CDKN2B, E2F5, MCM3, and MCM4. In summary our results showed an association between exposure to PCBs, phthalates, BPA, and increased risk of PAD and AS, and possible molecular mechanisms of interaction of ID3 target genes and estrogenic chemicals involved in PAD and AS.

Background: Meat consumption is associated with an elevated risk of colorectal cancer (CRC); exposure to heterocyclic aromatic amines (HAAs), carcinogens produced when meat is cooked at high temperatures, is one hypothesized explanation for this relationship. HAAs form adducts with DNA; left unrepaired, DNA adducts can induce mutations which may initiate and/or promote the development of colorectal adenomas, precursors to the vast majority of CRCs. Along this continuum, genetic differences in the ability to biotransform or metabolize HAAs and repair DNA is postulated to modify the HAA-CRC relationship. Methods: This thesis examined the HAA-CRC relationship in two studies (Phase 1 and 2). In a cross-sectional study of 99 healthy volunteers, Phase 1 investigated the relationship between dietary exposure to HAAs and the levels of bulky DNA adducts in blood leukocytes. In Phase 2, a cross-sectional study examined the relationships between dietary exposures to: a) HAAs and; b) meat mutagenicity, and the prevalence of colorectal adenomas among 342 patients undergoing a screening colonoscopy. Both Phase 1 and 2 examined potential gene-diet interactions between dietary HAAs and genetic factors relevant to the biotransformation of HAAs and DNA repair. Results: In Phase 1, an interaction was observed for dietary HAAs and NAT1 polymorphisms where a positive association between HAA intakes and bulky DNA adduct levels was found among those with the NAT1 slow acetylator genotype, hypothesized to confer a lower ability to biotransform HAAs. In Phase 2, polymorphisms in genes involved in the biotransformation of HAAs (CYP1B1 rs10012 and rs1056827) and DNA repair (XPC rs2228001) were found to determine colorectal adenoma risk. As well, gene-diet interactions were observed for dietary HAAs/meat mutagenicity exposures and polymorphisms in CYP1B1 and XPD (rs13181 and rs1799793). Overall, a higher risk of colorectal adenoma was observed with higher HAA and/or meat mutagenicity exposures among those with polymorphisms which confer a greater activity to biotransform HAAs and/or a lower ability to repair DNA. Conclusion: This research supports the contribution of dietary HAAs and genetic susceptibility to the risk of developing colorectal adenomas and highlighted bulky DNA adduct formation as a potential biologic pathway through which HAAs may influence cancer risk. / Thesis (Ph.D, Community Health & Epidemiology) -- Queen's University, 2014-04-25 11:32:30.392