Elizabeth D. Schifano

DNA Methylation Signatures Identify Biologically Distinct Subtypes in Acute Myeloid Leukemia

We hypothesized that DNA methylation distributes into specific patterns in cancer cells, which reflect critical biological differences. We therefore examined the methylation profiles of 344 patients with acute myeloid leukemia (AML). Clustering of these patients by methylation data segregated patients into 16 groups. Five of these groups defined new AML subtypes that shared no other known feature. In addition, DNA methylation profiles segregated patients with CEBPA aberrations from other subtypes of leukemia, defined four epigenetically distinct forms of AML with NPM1 mutations, and showed that established AML1-ETO, CBFb-MYH11, and PML-RARA leukemia entities are associated with specific methylation profiles. We report a 15 gene methylation classifier predictive of overall survival in an independent patient cohort (p < 0.001, adjusted for known covariates).

Genome-wide Association Analysis for Multiple Continuous Secondary Phenotypes

There is increasing interest in the joint analysis of multiple phenotypes in genome-wide association studies (GWASs), especially for the analysis of multiple secondary phenotypes in case-control studies and in detecting pleiotropic effects. Multiple phenotypes often measure the same underlying trait. By taking advantage of similarity across phenotypes, one could potentially gain statistical power in association analysis. Because continuous phenotypes are likely to be measured on different scales, we propose a scaled marginal model for testing and estimating the common effect of single-nucleotide polymorphism (SNP) on multiple secondary phenotypes in case-control studies. This approach improves power in comparison to individual phenotype analysis and traditional multivariate analysis when phenotypes are positively correlated and measure an underlying trait in the same direction (after transformation) by borrowing strength across outcomes with a one degree of freedom (1-DF) test and jointly estimating outcome-specific scales along with the SNP and covariate effects. To account for case-control ascertainment bias for the analysis of multiple secondary phenotypes, we propose weighted estimating equations for fitting scaled marginal models. This weighted estimating equation approach is robust to departures from normality of continuous multiple phenotypes and the misspecification of within-individual correlation among multiple phenotypes. Statistical power improves when the within-individual correlation is correctly specified. We perform simulation studies to show the proposed 1-DF common effect test outperforms several alternative methods. We apply the proposed method to investigate SNP associations with smoking behavior measured with multiple secondary smoking phenotypes in a lung cancer case-control GWAS and identify several SNPs of biological interest.

Majorization-Minimization algorithms for nonsmoothly penalized objective functions

The use of regularization, or penalization, has become incr easingly common in highdimensional statistical analysis over the past decade, whe re a common goal is to simultaneously select important variables and estimate their e ffects. It has been shown by several authors that these goals can be achieved by minimizing some parameter-depende nt “goodness of fit” function (e.g., a negative loglikelihood) subject to a penalization that pr omotes sparsity. Penalty functions that are nonsmooth (i.e. not di fferentiable) at the origin have received substantial attent ion, arguably beginning with LASSO (Tibshirani, 1996). The current literature tends to focus on specific combinatio s f smooth data fidelity (i.e., goodness-of-fit) and nonsmooth penalty functions. One resu lt of this combined specificity has been a proliferation in the number of computational algorithms d e igned to solve fairly narrow classes of optimization problems involving objective functions that are not everywhere continuously di fferentiable. In this paper, we propose a general class of algorith ms for optimizing an extensive variety of nonsmoothly penalized objective functions that satisfy ce rtain regularity conditions. The proposed framework utilizes the majorization-minimization (MM) al gorithm as its core optimization engine. The resulting algorithms rely on iterated soft-thresholdi ng, implemented componentwise, allowing for fast, stable updating that avoids the need for any high-d imensional matrix inversion. We establish a local convergence theory for this class of algorithms unde r weaker assumptions than previously considered in the statistical literature. We also demonstr ate he exceptional e ffectiveness of new acceleration methods, originally proposed for the EM algorit hm, in this class of problems. Simulation results and a microarray data example are provided to demons trate the algorithm’s capabilities and versatility.

Laplace Approximated EM Microarray Analysis: An Empirical Bayes Approach for Comparative Microarray Experiments

A two-groups mixed-effects model for the comparison of (normalized) microarray data from two treatment groups is considered. Most competing parametric methods that have appeared in the literature are obtained as special cases or by minor modification of the proposed model. Approximate maximum likelihood fitting is accomplished via a fast and scalable algorithm, which we call LEMMA (Laplace approximated EM Microarray Analysis). The posterior odds of treatment

Statistical methods and computing for big data.

\times

Online Updating of Statistical Inference in the Big Data Setting.

gene interactions, derived from the model, involve shrinkage estimates of both the interactions and of the gene specific error variances. Genes are classified as being associated with treatment based on the posterior odds and the local false discovery rate (f.d.r.) with a fixed cutoff. Our model-based approach also allows one to declare the non-null status of a gene by controlling the false discovery rate (FDR). It is shown in a detailed simulation study that the approach outperforms well-known competitors. We also apply the proposed methodology to two previously analyzed microarray examples. Extensions of the proposed method to paired treatments and multiple treatments are also discussed.

Hierarchical Bayes, maximum a posteriori estimators, and minimax concave penalized likelihood estimation

Big data are data on a massive scale in terms of volume, intensity, and complexity that exceed the capacity of standard analytic tools. They present opportunities as well as challenges to statisticians. The role of computational statisticians in scientific discovery from big data analyses has been under-recognized even by peer statisticians. This article summarizes recent methodological and software developments in statistics that address the big data challenges. Methodologies are grouped into three classes: subsampling-based, divide and conquer, and online updating for stream data. As a new contribution, the online updating approach is extended to variable selection with commonly used criteria, and their performances are assessed in a simulation study with stream data. Software packages are summarized with focuses on the open source R and R packages, covering recent tools that help break the barriers of computer memory and computing power. Some of the tools are illustrated in a case study with a logistic regression for the chance of airline delay.

Exposure to secondhand smoke and asthma severity among children in Connecticut

We present statistical methods for big data arising from online analytical processing, where large amounts of data arrive in streams and require fast analysis without storage/access to the historical data. In particular, we develop iterative estimating algorithms and statistical inferences for linear models and estimating equations that update as new data arrive. These algorithms are computationally efficient, minimally storage-intensive, and allow for possible rank deficiencies in the subset design matrices due to rare-event covariates. Within the linear model setting, the proposed online-updating framework leads to predictive residual tests that can be used to assess the goodness of fit of the hypothesized model. We also propose a new online-updating estimator under the estimating equation setting. Theoretical properties of the goodness-of-fit tests and proposed estimators are examined in detail. In simulation studies and real data applications, our estimator compares favorably with competing approaches under the estimating equation setting. Supplementary materials for this article are available online.

Differential variation and expression analysis

Priors constructed from scale mixtures of normal distributions have long played an important role in decision theory and shrinkage estimation. This paper demonstrates equivalence between the maximum aposteriori estimator constructed under one such prior and Zhang’s minimax concave penalization estimator. This equivalence and related multivariate generalizations stem directly from an intriguing representation of the minimax concave penalty function as the Moreau envelope of a simple convex function. Maximum aposteriori estimation under the corresponding marginal prior distribution, a generalization of the quasi-Cauchy distribution proposed by Johnstone and Silverman, leads to thresholding estimators having excellent frequentist risk properties. AMS 2000 subject classifications: Primary 62C60, 62J07.

Deep-targeted exon sequencing reveals renal polymorphisms associate with postexercise hypotension among African Americans

Objective To determine whether secondhand smoke (SHS) exposure is associated with greater asthma severity in children with physician-diagnosed asthma living in CT, and to examine whether area of residence, race/ethnicity or poverty moderate the association. Methods A large childhood asthma database in CT (Easy Breathing) was linked by participant zip code to census data to classify participants by area of residence. Multinomial logistic regression models, adjusted for enrollment date, sex, age, race/ethnicity, area of residence, insurance type, family history of asthma, eczema, and exposure to dogs, cats, gas stove, rodents and cockroaches were used to examine the association between self-reported exposure to SHS and clinician-determined asthma severity (mild, moderate, and severe persistent vs. intermittent asthma). Results Of the 30,163 children with asthma enrolled in Easy Breathing, between 6 months and 18 years old, living in 161 different towns in CT, exposure to SHS was associated with greater asthma severity (adjusted relative risk ratio (aRRR): 1.07 [1.00, 1.15] and aRRR: 1.11 [1.02, 1.22] for mild and moderate persistent asthma, respectively). The odds of Black and Puerto Rican/Hispanic children with asthma being exposed to SHS were twice that of Caucasian children. Though the odds of SHS exposure for publicly insured children with asthma were three times greater than the odds for privately insured children (OR: 3.02 [2.84,3,21]), SHS exposure was associated with persistent asthma only among privately insured children (adjusted odds ratio (aOR): 1.23 [1.11,1.37]). Conclusion This is the first large-scale pragmatic study to demonstrate that children exposed to SHS in Connecticut have greater asthma severity, clinically determined using a systematic approach, and varies by insurance status.