An empirical evaluation of the Random Forests classifier models for variable selection in a large-scale lung cancer case-control study

Random Forests™ is reported to be one of the most accurate classification algorithms in complex data analysis. It shows excellent performance even when most predictors are noisy and the number of variables is much larger than the number of observations. In this thesis Random Forests was applied to a large-scale lung cancer case-control study. A novel way of automatically selecting prognostic factors was proposed. Also, synthetic positive control was used to validate Random Forests method. Throughout this study we showed that Random Forests can deal with large number of weak input variables without overfitting. It can account for non-additive interactions between these input variables. Random Forests can also be used for variable selection without being adversely affected by collinearities. ^ Random Forests can deal with the large-scale data sets without rigorous data preprocessing. It has robust variable importance ranking measure. Proposed is a novel variable selection method in context of Random Forests that uses the data noise level as the cut-off value to determine the subset of the important predictors. This new approach enhanced the ability of the Random Forests algorithm to automatically identify important predictors for complex data. The cut-off value can also be adjusted based on the results of the synthetic positive control experiments. ^ When the data set had high variables to observations ratio, Random Forests complemented the established logistic regression. This study suggested that Random Forests is recommended for such high dimensionality data. One can use Random Forests to select the important variables and then use logistic regression or Random Forests itself to estimate the effect size of the predictors and to classify new observations. ^ We also found that the mean decrease of accuracy is a more reliable variable ranking measurement than mean decrease of Gini. ^

DEFINITION OF THE LANDSCAPE OF CHROMATIN STRUCTURE AT THE FRATAXIN GENE IN FRIEDREICH’S ATAXIA

Friedreich’s ataxia (FRDA) is caused by the transcriptional silencing of the frataxin (FXN) gene. FRDA patients have expansion of GAA repeats in intron 1 of the FXN gene in both alleles. A number of studies demonstrated that specific histone deacetylase inhibitors (HDACi) affect either histone modifications at the FXN gene or FXN expression in FRDA cells, indicating that the hyperexpanded GAA repeat may facilitate heterochromatin formation. However, the correlation between chromatin structure and transcription at the FXN gene is currently limited due to a lack of more detailed analysis. Therefore, I analyzed the effects of the hyperexpanded GAA repeats on transcription status and chromatin structure using lymphoid cell lines derived from FRDA patients. Using chromatin immunoprecipitation and quantitative PCR, I observed significant changes in the landscape of histone modifications in the vicinity of the GAA tract in FRDA cells relative to control cells. Similar epigenetic changes were observed in GFP reporter construct containing 560 GAA repeats. Further, I detected similar levels of FXN pre-mRNA at a region upstream of hyperexpanded GAA repeats in FRDA and control cells, indicating similar efficiency of transcription initiation in FRDA cells. I also showed that histone modifications associated with hyperexpanded GAA repeats are independent of transcription progression using the GFP reporter system. My data strongly support evidence that FXN deficiency in FRDA patients is consequence of defective transition from initiation to elongation of FXN transcription due to heterochromatin-like structures formed in the proximity of the hyperexpanded GAAs.

Structure-function relation in retinoid regulated gene expression

Retinoids are known to inhibit proliferation of and induce terminal differentiation of many normal and transformed cells. It has been postulated that retinoids exert their effect by altering gene expression. HL-60 cells and macrophages both respond to retinoic acid action by the rapid induction of the enzyme tissue transglutaminase. The induction has been shown to be due to increased transcription of the transglutaminase gene. The first part of the dissertation studied the structure-function relationship of retinoid-regulated transglutaminase induction, differentiation and proliferation in HL-60 cells using retinoid analogs. The results indicated strict structural constraints and a strong structure-function correlation between transglutaminase induction and differentiation; those retinoids that induced transglutaminase also induced differentiation, those analogs that did not induce transglutaminase could not induce differentiation. The ability of the retinoids to induce transglutaminase in HL-60 cells was paralleled in macrophages. However, the antiproliferative effect of the retinoids displayed less stringent structural constraints than their differentiation- and transglutaminase-inducing properties. Specifically all the retinoids were able to inhibit proliferation to varying extents. It is concluded that the induction of transglutaminase and of differentiation by retinoids is mediated by receptors. While receptor mediation cannot be entirely ruled out, with the current data no definitive statement can be made about the antiproliferative activity of retinoids. Also, the concordance in the ability of the retinoids to induce transglutaminase and the ability to induce differentiation of HL-60 cells suggests that the former is an early response of the cells to retinoids and differentiation a later consequence on the same pathway. Using the induction of transglutaminase as an index of the direct, or primary, effect of retinoids on gene expression, the second part of the dissertation investigates, by 2D gel electrophoresis, the alteration in the rates of synthesis of other proteins in macrophages and HL-60 cells in response to short incubations with retinoic acid. Any changes in parallel with transglutaminase were taken to indicate proteins directly under the control of retinoic acid. It is concluded that retinoic acid regulates the expression of a circumscribed set of genes in a cell-specific manner. The results support the hypothesis that retinoids exert their multiple effects on myeloid cells, in part, by receptor-mediated alternations in gene expression. ^

Molecular structure of antibodies against small ligands

Antibodies which bind bioactive ligands can serve as a template for the generation of a second antibody which may react with the physiological receptor. This phenomenon of molecular mimicry by antibodies has been described in a variety of systems. In order to understand the chemical and molecular mechanisms involved in these interactions, monoclonal antibodies directed against two pharmacologically active alkaloids, morphine and nicotine, were carefully studied using experimental and theoretical molecular modeling techniques. The molecular characterization of these antibodies involved binding studies with ligand analogs and determination of the variable region amino acid sequence. A three-dimensional model of the anti-morphine binding site was constructed using computational and graphics display techniques. The antibody response in BALB/c mice to morphine appears relatively restricted, in that all of the antibodies examined in this study contained a $\lambda$ light chain, which is normally found in only 5% of mouse immunoglobulins. This study represents the first use of theoretical and experimental modeling techniques to describe the antigen binding site of a mouse Fv region containing a $\lambda$ light chain. The binding site model indicates that a charged glutamic acid residue and aromatic side chains are key features in ionic and hydrophobic interactions with the ligand morphine. A glutamic acid residue is found in the identical position in the anti-nicotine antibody and may play a role in binding nicotine. ^

Effects of variable Gs expression on $\beta\sb2$-adrenergic receptor stimulated adenylyl cyclase response

An exact knowledge of the kinetic nature of the interaction between the stimulatory G protein (G$\sb{\rm s}$) and the adenylyl cyclase catalytic unit (C) is essential for interpreting the effects of Gs mutations and expression levels on cellular response to a wide variety of hormones, drugs, and neurotransmitters. In particular, insight as to the association of these proteins could lead to progress in tumor biology where single spontaneous mutations in G proteins have been associated with the formation of tumors (118). The question this work attempts to answer is whether the adenylyl cyclase activation by epinephrine stimulated $\beta\sb2$-adrenergic receptors occurs via G$\sb{\rm s}$ proteins by a G$\sb{\rm s}$ to C shuttle or G$\sb{\rm s}$-C precoupled mechanism. The two forms of activation are distinguishable by the effect of G$\sb{\rm s}$ levels on epinephrine stimulated EC50 values for cyclase activation.^ We have made stable transfectants of S49 cyc$\sp-$ cells with the gene for the $\alpha$ protein of G$\sb{\rm s}$ $(\alpha\sb{\rm s})$ which is under the control of the mouse mammary tumor virus LTR promoter (110). Expression of G$\sb{\rm s}\alpha$ was then controlled by incubation of the cells for various times with 5 $\mu$M dexamethasone. Expression of G$\sb{\rm s}\alpha$ led to the appearance of GTP shifts in the competitive binding of epinephrine with $\sp{125}$ICYP to the $\beta$-adrenergic receptors and to agonist dependent adenylyl cyclase activity. High expression of G$\sb{\rm s}\alpha$ resulted in lower EC50's for the adenylyl cyclase activity in response to epinephrine than did low expression. By kinetic modelling, this result is consistent with the existence of a shuttle mechanism for adenylyl cyclase activation by hormones.^ One item of concern that remains to be addressed is the extent to which activation of adenylyl cyclase occurs by a "pure" shuttle mechanism. Kinetic and biochemical experiments by other investigators have revealed that adenylyl cyclase activation, by hormones, may occur via a Gs-C precoupled mechanism (80, 94, 97). Activation of adenylyl cyclase, therefore, probably does not occur by either a pure "'Shuttle" or "Gs-C Precoupled" mechanism, but rather by a "Hybrid" mechanism. The extent to which either the shuttle or precoupled mechanism contributes to hormone stimulated adenylyl cyclase activity is the subject of on-going research. ^

Genes to blood pressure: The effects of variation in genes of the renin-angiotensin system on the hormonal and renal control of blood pressure

Objective. Essential hypertension affects 25% of the US adult population and is a leading contributor to morbidity and mortality. Because BP is a multifactorial phenotype that resists simple genetic analysis, intermediate phenotypes within the complex network of BP regulatory systems may be more accessible to genetic dissection. The Renin-Angiotensin System (RAS) is known to influence intermediate and long-term blood pressure regulation through alterations in vascular tone and renal sodium and fluid resorption. This dissertation examines associations between renin (REN), angiotensinogen (AGT), angiotensin-converting enzyme (ACE) and angiotensin II type 1 receptor (AT1) gene variation and interindividual differences in plasma hormone levels, renal hemodynamics, and BP homeostasis.^ Methods. A total of 150 unrelated men and 150 unrelated women, between 20.0 and 49.9 years of age and free of acute or chronic illness except for a history of hypertension (11 men and 7 women, all off medications), were studied after one week on a controlled sodium diet. RAS plasma hormone levels, renal hemodynamics and BP were determined prior to and during angiotensin II (Ang II) infusion. Individuals were genotyped by PCR for a variable number tandem repeat (VNTR) polymorphism in REN, and for the following restriction fragment length polymorphisms (RFLP): AGT M235T, ACE I/D, and AT1 A1166C. Associations between clinical measurements and allelic variation were examined using multiple linear regression statistical models.^ Results. Women homozygous for the AT1 1166C allele demonstrated higher intracellular levels of sodium (p = 0.044). Men homozygous for the AGT T235 allele demonstrated a blunted decrement in renal plasma flow in response to Ang II infusion (p = 0.0002). There were no significant associations between RAS gene variation and interindividual variation in RAS plasma hormone levels or BP.^ Conclusions. Rather than identifying new BP controlling genes or alleles, the study paradigm employed in this thesis (i.e., measured genes, controlled environments and interventions) may provide mechanistic insight into how candidate genes affect BP homeostasis. ^

Multiple imputation for missing continuous and dichotomous data: The effects of patterns of missingness and variable correlations on type I error with small samples

The purpose of this study is to investigate the effects of predictor variable correlations and patterns of missingness with dichotomous and/or continuous data in small samples when missing data is multiply imputed. Missing data of predictor variables is multiply imputed under three different multivariate models: the multivariate normal model for continuous data, the multinomial model for dichotomous data and the general location model for mixed dichotomous and continuous data. Subsequent to the multiple imputation process, Type I error rates of the regression coefficients obtained with logistic regression analysis are estimated under various conditions of correlation structure, sample size, type of data and patterns of missing data. The distributional properties of average mean, variance and correlations among the predictor variables are assessed after the multiple imputation process. ^ For continuous predictor data under the multivariate normal model, Type I error rates are generally within the nominal values with samples of size n = 100. Smaller samples of size n = 50 resulted in more conservative estimates (i.e., lower than the nominal value). Correlation and variance estimates of the original data are retained after multiple imputation with less than 50% missing continuous predictor data. For dichotomous predictor data under the multinomial model, Type I error rates are generally conservative, which in part is due to the sparseness of the data. The correlation structure for the predictor variables is not well retained on multiply-imputed data from small samples with more than 50% missing data with this model. For mixed continuous and dichotomous predictor data, the results are similar to those found under the multivariate normal model for continuous data and under the multinomial model for dichotomous data. With all data types, a fully-observed variable included with variables subject to missingness in the multiple imputation process and subsequent statistical analysis provided liberal (larger than nominal values) Type I error rates under a specific pattern of missing data. It is suggested that future studies focus on the effects of multiple imputation in multivariate settings with more realistic data characteristics and a variety of multivariate analyses, assessing both Type I error and power. ^

Comparing the factorial structure, invariance, and predictive validity of transtheoretical model constructs for alcohol use across restricted and unrestricted settings

With substance abuse treatment expanding in prisons and jails, understanding how behavior change interacts with a restricted setting becomes more essential. The Transtheoretical Model (TTM) has been used to understand intentional behavior change in unrestricted settings, however, evidence indicates restrictive settings can affect the measurement and structure of the TTM constructs. The present study examined data from problem drinkers at baseline and end-of-treatment from three studies: (1) Project CARE (n = 187) recruited inmates from a large county jail; (2) Project Check-In (n = 116) recruited inmates from a state prison; (3) Project MATCH, a large multi-site alcohol study had two recruitment arms, aftercare (n = 724 pre-treatment and 650 post-treatment) and outpatient (n = 912 pre-treatment and 844 post-treatment). The analyses were conducted using cross-sectional data to test for non-invariance of measures of the TTM constructs: readiness, confidence, temptation, and processes of change (Structural Equation Modeling, SEM) across restricted and unrestricted settings. Two restricted (jail and aftercare) and one unrestricted group (outpatient) entering treatment and one restricted (prison) and two unrestricted groups (aftercare and outpatient) at end-of-treatment were contrasted. In addition TTM end-of-treatment profiles were tested as predictors of 12 month drinking outcomes (Profile Analysis). Although SEM did not indicate structural differences in the overall TTM construct model across setting types, there were factor structure differences on the confidence and temptation constructs at pre-treatment and in the factor structure of the behavioral processes at the end-of-treatment. For pre-treatment temptation and confidence, differences were found in the social situations factor loadings and in the variance for the confidence and temptation latent factors. For the end-of-treatment behavioral processes, differences across the restricted and unrestricted settings were identified in the counter-conditioning and stimulus control factor loadings. The TTM end-of-treatment profiles were not predictive of drinking outcomes in the prison sample. Both pre and post-treatment differences in structure across setting types involved constructs operationalized with behaviors that are limited for those in restricted settings. These studies suggest the TTM is a viable model for explicating addictive behavior change in restricted settings but calls for modification of subscale items that refer to specific behaviors and caution in interpreting the mean differences across setting types for problem drinkers. ^

Does parental monitoring influence the use of alcohol and drugs among inner city 7th grade students?

Objective. To examine associations between parental monitoring and adolescent alcohol/drug use. ^ Methods. 981 7th grade students from 10 inner-city middle schools were surveyed at the 3 month follow-up of an HIV, STD, and pregnancy prevention program. Data from 549 control subjects were used for analyses. Multinomial logistic regression was used to examine associations between five parental monitoring variables and substance use, coded as: low risk [never drank alcohol or used drugs (0)], moderate risk [drank alcohol, no drug use (1)], and high risk [both drank alcohol and used drugs or just used drugs (2)]. ^ Results. Participants were 58.3% female, 39.6% African American, 43.8% Hispanic, mean age 13.3 years. Lifetime alcohol use was 47.9%. Lifetime drug use was 14.9%. Adjusted for gender, age, race, and family structure, each individual parental monitoring variable (perceived parental monitoring, less permissive parental monitoring, greater supervision (public places), greater supervision (teen clubs), and less time spent with older teens) was significant and protective for the moderate and high risk groups. When all 5 variables were entered into a single model, only perceived parental monitoring was significantly associated (OR=0.40, 95% CI 0.29-0.55) for the moderate risk group. For the high risk group, 3 variables were significantly protective (perceived parental monitoring OR=0.28, CI 0.18-0.42, less time spent with older teens OR=0.75, CI 0.60-0.93, and greater supervision (public places) OR=0.79, CI 0.64-0.99). ^ Conclusion. The association between parental monitoring and substance abuse is complex and varied for different risk levels. Implications for intervention development are addressed. ^

Host risk factors for rifamycin-resistant clostridium difficile infection: A case control study

Objective. To evaluate the host risk factors associated with rifamycin-resistant Clostridium difficile (C. diff) infection in hospitalized patients compared to rifamycin-susceptible C.diff infection.^ Background. C. diff is the most common definable cause of nosocomial diarrhea affecting elderly hospitalized patients taking antibiotics for prolonged durations. The epidemiology of Clostridium difficile associated disease is now changing with the reports of a new hypervirulent strain causing hospital outbreaks. This new strain is associated with increased disease severity and mortality. The conventional therapy for C. diff includes metronidazole and vancomycin but high recurrence rates and treatment failures are now becoming a major concern. Rifamycin antibiotics are being developed as a new therapeutic option to treat C. diff infection after their efficacy was established in a few in vivo and in vitro studies. There are some recent studies that report an association between the hypervirulent strain and emerging rifamycin resistance. These findings assess the need for clinical studies to better understand the efficacy of rifamycin drugs against C. diff.^ Methods. This is a hospital-based, matched case-control study using de-identified data drawn from two prospective cohort studies involving C. diff patients at St Luke's Hospital. The C. diff isolates from these patients are screened for rifamycin resistance using agar dilution methods for minimum inhibitory concentrations (MIC) as part of Dr Zhi-Dong Jiang's study. Twenty-four rifamycin-rifamycin resistant C. diff cases were identified and matched with one rifamycin susceptible C. diff control on the basis of ± 10 years of age and hospitalization 30 days before or after the case. De-identified data for the 48 subjects was obtained from Dr Kevin Garey's clinical study at St Luke's Hospital enrolling C. diff patients. It was reviewed to gather information about host risk factors, outcome variables and relevant clinical characteristic.^ Results. Medical diagnosis at the time of admission (p = 0.0281) and history of chemotherapy (p = 0.022) were identified as a significant risk factor while hospital stay ranging from 1 week to 1 month and artificial feeding were identified as an important outcome variable (p = 0.072 and p = 0.081 respectively). Horn's Index assessing the severity of underlying illness and duration of antibiotics for cases and controls showed no significant difference.^ Conclusion. The study was a small project designed to identify host risk factors and understand the clinical implications of rifamycin-resistance. The study was underpowered and a larger sample size is needed to validate the results.^

Multivariate methods for correlated effects sizes

Current statistical methods for estimation of parametric effect sizes from a series of experiments are generally restricted to univariate comparisons of standardized mean differences between two treatments. Multivariate methods are presented for the case in which effect size is a vector of standardized multivariate mean differences and the number of treatment groups is two or more. The proposed methods employ a vector of independent sample means for each response variable that leads to a covariance structure which depends only on correlations among the $p$ responses on each subject. Using weighted least squares theory and the assumption that the observations are from normally distributed populations, multivariate hypotheses analogous to common hypotheses used for testing effect sizes were formulated and tested for treatment effects which are correlated through a common control group, through multiple response variables observed on each subject, or both conditions.^ The asymptotic multivariate distribution for correlated effect sizes is obtained by extending univariate methods for estimating effect sizes which are correlated through common control groups. The joint distribution of vectors of effect sizes (from $p$ responses on each subject) from one treatment and one control group and from several treatment groups sharing a common control group are derived. Methods are given for estimation of linear combinations of effect sizes when certain homogeneity conditions are met, and for estimation of vectors of effect sizes and confidence intervals from $p$ responses on each subject. Computational illustrations are provided using data from studies of effects of electric field exposure on small laboratory animals. ^

HEAD OF HOUSEHOLD STRUCTURE: ITS INFLUENCE ON THE DIETARY QUALITY OF U.S. HOUSEHOLDS, 1977-78 (UNITED STATES, NATIONWIDE FOOD CONSUMPTION SURVEY)

Using analysis of variance, household data collected in the Spring portion of the 1977-78 Nationwide Food Consumption Survey conducted by the United States Department of Agriculture were analyzed to examine the relationship between household characteristics and dietary quality of the household food supply. Results indicated that head of household structure was a statistically significant variable, with female headed households having higher dietary quality.^ Further analysis indicated that neither race, degree of urbanization, regional location, the education level of the female head, nor her employment status were significant variables in influencing dietary quality. The influence of head of household structure remained significant when these variables were controlled. However, income, household size, and family life cycle stage had statistically significant effects on dietary quality, and when individually controlled, the influence of head of household structure disappeared. ^

BAYESIAN STATISTICAL METHODS IN GENE-ENVIRONMENT AND GENE-GENE INTERACTION STUDIES

Complex diseases such as cancer result from multiple genetic changes and environmental exposures. Due to the rapid development of genotyping and sequencing technologies, we are now able to more accurately assess causal effects of many genetic and environmental factors. Genome-wide association studies have been able to localize many causal genetic variants predisposing to certain diseases. However, these studies only explain a small portion of variations in the heritability of diseases. More advanced statistical models are urgently needed to identify and characterize some additional genetic and environmental factors and their interactions, which will enable us to better understand the causes of complex diseases. In the past decade, thanks to the increasing computational capabilities and novel statistical developments, Bayesian methods have been widely applied in the genetics/genomics researches and demonstrating superiority over some regular approaches in certain research areas. Gene-environment and gene-gene interaction studies are among the areas where Bayesian methods may fully exert its functionalities and advantages. This dissertation focuses on developing new Bayesian statistical methods for data analysis with complex gene-environment and gene-gene interactions, as well as extending some existing methods for gene-environment interactions to other related areas. It includes three sections: (1) Deriving the Bayesian variable selection framework for the hierarchical gene-environment and gene-gene interactions; (2) Developing the Bayesian Natural and Orthogonal Interaction (NOIA) models for gene-environment interactions; and (3) extending the applications of two Bayesian statistical methods which were developed for gene-environment interaction studies, to other related types of studies such as adaptive borrowing historical data. We propose a Bayesian hierarchical mixture model framework that allows us to investigate the genetic and environmental effects, gene by gene interactions (epistasis) and gene by environment interactions in the same model. It is well known that, in many practical situations, there exists a natural hierarchical structure between the main effects and interactions in the linear model. Here we propose a model that incorporates this hierarchical structure into the Bayesian mixture model, such that the irrelevant interaction effects can be removed more efficiently, resulting in more robust, parsimonious and powerful models. We evaluate both of the 'strong hierarchical' and 'weak hierarchical' models, which specify that both or one of the main effects between interacting factors must be present for the interactions to be included in the model. The extensive simulation results show that the proposed strong and weak hierarchical mixture models control the proportion of false positive discoveries and yield a powerful approach to identify the predisposing main effects and interactions in the studies with complex gene-environment and gene-gene interactions. We also compare these two models with the 'independent' model that does not impose this hierarchical constraint and observe their superior performances in most of the considered situations. The proposed models are implemented in the real data analysis of gene and environment interactions in the cases of lung cancer and cutaneous melanoma case-control studies. The Bayesian statistical models enjoy the properties of being allowed to incorporate useful prior information in the modeling process. Moreover, the Bayesian mixture model outperforms the multivariate logistic model in terms of the performances on the parameter estimation and variable selection in most cases. Our proposed models hold the hierarchical constraints, that further improve the Bayesian mixture model by reducing the proportion of false positive findings among the identified interactions and successfully identifying the reported associations. This is practically appealing for the study of investigating the causal factors from a moderate number of candidate genetic and environmental factors along with a relatively large number of interactions. The natural and orthogonal interaction (NOIA) models of genetic effects have previously been developed to provide an analysis framework, by which the estimates of effects for a quantitative trait are statistically orthogonal regardless of the existence of Hardy-Weinberg Equilibrium (HWE) within loci. Ma et al. (2012) recently developed a NOIA model for the gene-environment interaction studies and have shown the advantages of using the model for detecting the true main effects and interactions, compared with the usual functional model. In this project, we propose a novel Bayesian statistical model that combines the Bayesian hierarchical mixture model with the NOIA statistical model and the usual functional model. The proposed Bayesian NOIA model demonstrates more power at detecting the non-null effects with higher marginal posterior probabilities. Also, we review two Bayesian statistical models (Bayesian empirical shrinkage-type estimator and Bayesian model averaging), which were developed for the gene-environment interaction studies. Inspired by these Bayesian models, we develop two novel statistical methods that are able to handle the related problems such as borrowing data from historical studies. The proposed methods are analogous to the methods for the gene-environment interactions on behalf of the success on balancing the statistical efficiency and bias in a unified model. By extensive simulation studies, we compare the operating characteristics of the proposed models with the existing models including the hierarchical meta-analysis model. The results show that the proposed approaches adaptively borrow the historical data in a data-driven way. These novel models may have a broad range of statistical applications in both of genetic/genomic and clinical studies.

Instrumental variable method for the causal relationship between soluble receptor for advanced glycation end-products (sRAGE) and risk of pancreatic cancer

This thesis project is motivated by the potential problem of using observational data to draw inferences about a causal relationship in observational epidemiology research when controlled randomization is not applicable. Instrumental variable (IV) method is one of the statistical tools to overcome this problem. Mendelian randomization study uses genetic variants as IVs in genetic association study. In this thesis, the IV method, as well as standard logistic and linear regression models, is used to investigate the causal association between risk of pancreatic cancer and the circulating levels of soluble receptor for advanced glycation end-products (sRAGE). Higher levels of serum sRAGE were found to be associated with a lower risk of pancreatic cancer in a previous observational study (255 cases and 485 controls). However, such a novel association may be biased by unknown confounding factors. In a case-control study, we aimed to use the IV approach to confirm or refute this observation in a subset of study subjects for whom the genotyping data were available (178 cases and 177 controls). Two-stage IV method using generalized method of moments-structural mean models (GMM-SMM) was conducted and the relative risk (RR) was calculated. In the first stage analysis, we found that the single nucleotide polymorphism (SNP) rs2070600 of the receptor for advanced glycation end-products (AGER) gene meets all three general assumptions for a genetic IV in examining the causal association between sRAGE and risk of pancreatic cancer. The variant allele of SNP rs2070600 of the AGER gene was associated with lower levels of sRAGE, and it was neither associated with risk of pancreatic cancer, nor with the confounding factors. It was a potential strong IV (F statistic = 29.2). However, in the second stage analysis, the GMM-SMM model failed to converge due to non- concaveness probably because of the small sample size. Therefore, the IV analysis could not support the causality of the association between serum sRAGE levels and risk of pancreatic cancer. Nevertheless, these analyses suggest that rs2070600 was a potentially good genetic IV for testing the causality between the risk of pancreatic cancer and sRAGE levels. A larger sample size is required to conduct a credible IV analysis.^