Charles Kooperberg

Logspline Density Estimation for Censored Data

Abstract Logspline density estimation is developed for data that may be right censored, left censored, or interval censored. A fully automatic method, which involves the maximum likelihood method and may involve stepwise knot deletion and either the Akaike information criterion (AIC) or Bayesian information criterion (BIC), is used to determine the estimate. In solving the maximum likelihood equations, the Newton–Raphson method is augmented by occasional searches in the direction of steepest ascent. Also, a user interface based on S is described for obtaining estimates of the density function, distribution function, and quantile function and for generating a random sample from the fitted distribution.

Meta-analysis of gene-level tests for rare variant association.

The majority of reported complex disease associations for common genetic variants have been identified through meta-analysis, a powerful approach that enables the use of large sample sizes while protecting against common artifacts due to population structure and repeated small-sample analyses sharing individual-level data. As the focus of genetic association studies shifts to rare variants, genes and other functional units are becoming the focus of analysis. Here we propose and evaluate new approaches for performing meta-analysis of rare variant association tests, including burden tests, weighted burden tests, variable-threshold tests and tests that allow variants with opposite effects to be grouped together. We show that our approach retains useful features from single-variant meta-analysis approaches and demonstrate its use in a study of blood lipid levels in ∼18,500 individuals genotyped with exome arrays.

Characterization of Gene–Environment Interactions for Colorectal Cancer Susceptibility Loci

Genome-wide association studies (GWAS) have identified more than a dozen loci associated with colorectal cancer (CRC) risk. Here, we examined potential effect-modification between single-nucleotide polymorphisms (SNP) at 10 of these loci and probable or established environmental risk factors for CRC in 7,016 CRC cases and 9,723 controls from nine cohort and case-control studies. We used meta-analysis of an efficient empirical-Bayes estimator to detect potential multiplicative interactions between each of the SNPs [rs16892766 at 8q23.3 (EIF3H/UTP23), rs6983267 at 8q24 (MYC), rs10795668 at 10p14 (FLJ3802842), rs3802842 at 11q23 (LOC120376), rs4444235 at 14q22.2 (BMP4), rs4779584 at 15q13 (GREM1), rs9929218 at 16q22.1 (CDH1), rs4939827 at 18q21 (SMAD7), rs10411210 at 19q13.1 (RHPN2), and rs961253 at 20p12.3 (BMP2)] and select major CRC risk factors (sex, body mass index, height, smoking status, aspirin/nonsteroidal anti-inflammatory drug use, alcohol use, and dietary intake of calcium, folate, red meat, processed meat, vegetables, fruit, and fiber). The strongest statistical evidence for a gene-environment interaction across studies was for vegetable consumption and rs16892766, located on chromosome 8q23.3, near the EIF3H and UTP23 genes (nominal P(interaction) = 1.3 × 10(-4); adjusted P = 0.02). The magnitude of the main effect of the SNP increased with increasing levels of vegetable consumption. No other interactions were statistically significant after adjusting for multiple comparisons. Overall, the association of most CRC susceptibility loci identified in initial GWAS seems to be invariant to the other risk factors considered; however, our results suggest potential modification of the rs16892766 effect by vegetable consumption.

Inflammation and Thrombosis Biomarkers and Incident Frailty in Postmenopausal Women

BACKGROUND The immune and blood coagulation systems have been implicated in the pathophysiology of the geriatric syndrome of frailty, but limited prospective data examining the relationship of clotting/inflammation biomarkers to risk of incident frailty exist. METHODS This prospective analysis was derived from a nested case-control study within the Womens Health Initiative. Among women 65 to 79 years free of frailty at enrollment, we randomly selected 900 incident cases from those developing frailty within 3 years; 900 non-frail controls were individually matched on age, ethnicity, and blood collection date. Biomarkers assessed for risk of incident frailty included fibrinogen, factor VIII, D-dimer, C-reactive protein, interleukin-6, and tissue plasminogen activator (t-PA). RESULTS When examined by quartiles in multivariable adjusted models, higher D-dimer and t-PA levels were each associated with increased risk of frailty (P trend = .04). Relative to the lowest quartile, the odds ratios for frailty compared with the upper quartile were 1.52 (95% confidence interval, 1.05-2.22) for t-PA and 1.57 (95% confidence interval, 1.11-2.22) for D-dimer. For women having high t-PA and high D-dimer compared with women having lower levels of both biomarkers, the odds of frailty was 2.20 (1.29-3.75). There was little evidence for association between coagulation factor VIII, fibrinogen, C-reactive protein, or interleukin-6 levels and incident frailty. CONCLUSION This prospective analysis supports the role of markers of fibrin turnover and fibrinolysis as independent predictors of incident frailty in postmenopausal women.

Associations of Serum Sex Hormone-Binding Globulin and Sex Hormone Concentrations with Hip Fracture Risk in Postmenopausal Women

CONTEXT Endogenous estradiol, testosterone, and SHBG may influence the risk of hip fracture. DESIGN AND METHODS From the Womens Health Initiative Observational Study, 39,793 eligible postmenopausal women did not have a previous hip fracture and were not using estrogen or other bone-active therapies. Of these, 400 who had a first-time nonpathological hip fracture (median follow-up, 7 yr) were matched to 400 controls by age, ethnicity, and baseline blood draw date. Estradiol, testosterone, and SHBG were measured in banked baseline serum. RESULTS Compared with women in the lowest tertiles, those with bioavailable testosterone in the highest tertile had a lower risk [odds ratio (OR) = 0.62; 95% confidence interval (CI) = 0.44-0.88]; those with bioavailable estradiol in the highest tertile had a lower risk (OR = 0.44; 95% CI = 0.29-0.66), and those with SHBG in the highest tertile had a higher risk (OR = 1.90; 95% CI = 1.31-2.74) of hip fracture. In models with all three hormones and potential confounders, high SHBG remained a strong independent risk factor (OR = 1.76; 95% CI = 1.12-2.78), high bioavailable testosterone remained protective (OR = 0.64; 95% CI = 0.40-1.00), but estradiol no longer was associated (OR = 0.72; 95% CI = 0.42-1.23). CONCLUSIONS High serum SHBG is associated with an increased risk of subsequent hip fracture and high endogenous testosterone with a decreased risk, independent of each other, serum estradiol concentration, and other putative risk factors. But endogenous estradiol has no independent association with hip fracture.

The Cross‐Sectional Relationship Between Body Mass Index, Waist–Hip Ratio, and Cognitive Performance in Postmenopausal Women Enrolled in the Women's Health Initiative

OBJECTIVES: To determine whether body mass index (BMI) is independently associated with cognitive function in postmenopausal women and the relationship between body fat distribution as estimated by waist‐hip ratio (WHR).

Variation in 24 hemostatic genes and associations with non‐fatal myocardial infarction and ischemic stroke

Summary.  Background:  Arterial thrombosis involves platelet aggregation and clot formation, yet little is known about the contribution of genetic variation in fibrin‐based hemostatic factors to arterial clotting risk. We hypothesized that common variation in 24 coagulation–fibrinolysis genes would contribute to risk of incident myocardial infarction (MI) or ischemic stroke (IS).

Trees and splines in survival analysis

During the past few years several nonparametric alternatives to the Cox proportional hazards model have appeared in the literature. These methods extend techniques that are well known from regression analysis to the analysis of censored survival data. In this paper we discuss methods based on (partition) trees and (polynomial) splines, analyse two datasets using both Survival Trees and HARE, and compare the strengths and weaknesses of the two methods. One of the strengths of HARE is that its model fitting procedure has an implicit check for proportionality of the underlying hazards model. It also provides an explicit model for the conditional hazards function, which makes it very convenient to obtain graphical summaries. On the other hand, the tree-based methods automatically partition a dataset into groups of cases that are similar in survival history. Results obtained by survival trees and HARE are often complementary. Trees and splines in survival analysis should provide the data analyst with two useful tools when analysing survival data.

Semiparametric Estimation Exploiting Covariate Independence in Two‐Phase Randomized Trials

Recent results for case-control sampling suggest when the covariate distribution is constrained by gene-environment independence, semiparametric estimation exploiting such independence yields a great deal of efficiency gain. We consider the efficient estimation of the treatment-biomarker interaction in two-phase sampling nested within randomized clinical trials, incorporating the independence between a randomized treatment and the baseline markers. We develop a Newton-Raphson algorithm based on the profile likelihood to compute the semiparametric maximum likelihood estimate (SPMLE). Our algorithm accommodates both continuous phase-one outcomes and continuous phase-two biomarkers. The profile information matrix is computed explicitly via numerical differentiation. In certain situations where computing the SPMLE is slow, we propose a maximum estimated likelihood estimator (MELE), which is also capable of incorporating the covariate independence. This estimated likelihood approach uses a one-step empirical covariate distribution, thus is straightforward to maximize. It offers a closed-form variance estimate with limited increase in variance relative to the fully efficient SPMLE. Our results suggest exploiting the covariate independence in two-phase sampling increases the efficiency substantially, particularly for estimating treatment-biomarker interactions.

Whole-exome imputation of sequence variants identified two novel alleles associated with adult body height in African Americans

Adult body height is a quantitative trait for which genome-wide association studies (GWAS) have identified numerous loci, primarily in European populations. These loci, comprising common variants, explain <10% of the phenotypic variance in height. We searched for novel associations between height and common (minor allele frequency, MAF ≥5%) or infrequent (0.5% < MAF < 5%) variants across the exome in African Americans. Using a reference panel of 1692 African Americans and 471 Europeans from the National Heart, Lung, and Blood Institutes (NHLBI) Exome Sequencing Project (ESP), we imputed whole-exome sequence data into 13 719 African Americans with existing array-based GWAS data (discovery). Variants achieving a height-association threshold of P < 5E-06 in the imputed dataset were followed up in an independent sample of 1989 African Americans with whole-exome sequence data (replication). We used P < 2.5E-07 (=0.05/196 779 variants) to define statistically significant associations in meta-analyses combining the discovery and replication sets (N = 15 708). We discovered and replicated three independent loci for association: 5p13.3/C5orf22/rs17410035 (MAF = 0.10, β = 0.64 cm, P = 8.3E-08), 13q14.2/SPRYD7/rs114089985 (MAF = 0.03, β = 1.46 cm, P = 4.8E-10) and 17q23.3/GH2/rs2006123 (MAF = 0.30; β = 0.47 cm; P = 4.7E-09). Conditional analyses suggested 5p13.3 (C5orf22/rs17410035) and 13q14.2 (SPRYD7/rs114089985) may harbor novel height alleles independent of previous GWAS-identified variants (r(2) with GWAS loci <0.01); whereas 17q23.3/GH2/rs2006123 was correlated with GWAS-identified variants in European and African populations. Notably, 13q14.2/rs114089985 is infrequent in African Americans (MAF = 3%), extremely rare in European Americans (MAF = 0.03%), and monomorphic in Asian populations, suggesting it may be an African-American-specific height allele. Our findings demonstrate that whole-exome imputation of sequence variants can identify low-frequency variants and discover novel variants in non-European populations.