Projected t-SNE for batch correction.

MOTIVATION: Low-dimensional representations of high-dimensional data are routinely employed in biomedical research to visualize, interpret and communicate results from different pipelines. In this article, we propose a novel procedure to directly estimate t-SNE embeddings that are not driven by batch effects. Without correction, interesting structure in the data can be obscured by batch effects. The proposed algorithm can therefore significantly aid visualization of high-dimensional data. RESULTS: The proposed methods are based on linear algebra and constrained optimization, leading to efficient algorithms and fast computation in many high-dimensional settings. Results on artificial single-cell transcription profiling data show that the proposed procedure successfully removes multiple batch effects from t-SNE embeddings, while retaining fundamental information on cell types. When applied to single-cell gene expression data to investigate mouse medulloblastoma, the proposed method successfully removes batches related with mice identifiers and the date of the experiment, while preserving clusters of oligodendrocytes, astrocytes, and endothelial cells and microglia, which are expected to lie in the stroma within or adjacent to the tumours. AVAILABILITY AND IMPLEMENTATION: Source code implementing the proposed approach is available as an R package at https://github.com/emanuelealiverti/BC_tSNE, including a tutorial to reproduce the simulation studies. CONTACT: aliverti@stat.unipd.it.

Efficient posterior sampling for high-dimensional imbalanced logistic regression.

Classification with high-dimensional data is of widespread interest and often involves dealing with imbalanced data. Bayesian classification approaches are hampered by the fact that current Markov chain Monte Carlo algorithms for posterior computation become inefficient as the number [Formula: see text] of predictors or the number [Formula: see text] of subjects to classify gets large, because of the increasing computational time per step and worsening mixing rates. One strategy is to employ a gradient-based sampler to improve mixing while using data subsamples to reduce the per-step computational complexity. However, the usual subsampling breaks down when applied to imbalanced data. Instead, we generalize piecewise-deterministic Markov chain Monte Carlo algorithms to include importance-weighted and mini-batch subsampling. These maintain the correct stationary distribution with arbitrarily small subsamples and substantially outperform current competitors. We provide theoretical support for the proposed approach and demonstrate its performance gains in simulated data examples and an application to cancer data.

The whole-genome landscape of Burkitt lymphoma subtypes.

Burkitt lymphoma (BL) is an aggressive, MYC-driven lymphoma comprising 3 distinct clinical subtypes: sporadic BLs that occur worldwide, endemic BLs that occur predominantly in sub-Saharan Africa, and immunodeficiency-associated BLs that occur primarily in the setting of HIV. In this study, we comprehensively delineated the genomic basis of BL through whole-genome sequencing (WGS) of 101 tumors representing all 3 subtypes of BL to identify 72 driver genes. These data were additionally informed by CRISPR screens in BL cell lines to functionally annotate the role of oncogenic drivers. Nearly every driver gene was found to have both coding and non-coding mutations, highlighting the importance of WGS for identifying driver events. Our data implicate coding and non-coding mutations in IGLL5, BACH2, SIN3A, and DNMT1. Epstein-Barr virus (EBV) infection was associated with higher mutation load, with type 1 EBV showing a higher mutational burden than type 2 EBV. Although sporadic and immunodeficiency-associated BLs had similar genetic profiles, endemic BLs manifested more frequent mutations in BCL7A and BCL6 and fewer genetic alterations in DNMT1, SNTB2, and CTCF. Silencing mutations in ID3 were a common feature of all 3 subtypes of BL. In vitro, mass spectrometry-based proteomics demonstrated that the ID3 protein binds primarily to TCF3 and TCF4. In vivo knockout of ID3 potentiated the effects of MYC, leading to rapid tumorigenesis and tumor phenotypes consistent with those observed in the human disease.

Tensor network factorizations: Relationships between brain structural connectomes and traits.

Advanced brain imaging techniques make it possible to measure individuals' structural connectomes in large cohort studies non-invasively. Given the availability of large scale data sets, it is extremely interesting and important to build a set of advanced tools for structural connectome extraction and statistical analysis that emphasize both interpretability and predictive power. In this paper, we developed and integrated a set of toolboxes, including an advanced structural connectome extraction pipeline and a novel tensor network principal components analysis (TN-PCA) method, to study relationships between structural connectomes and various human traits such as alcohol and drug use, cognition and motion abilities. The structural connectome extraction pipeline produces a set of connectome features for each subject that can be organized as a tensor network, and TN-PCA maps the high-dimensional tensor network data to a lower-dimensional Euclidean space. Combined with classical hypothesis testing, canonical correlation analysis and linear discriminant analysis techniques, we analyzed over 1100 scans of 1076 subjects from the Human Connectome Project (HCP) and the Sherbrooke test-retest data set, as well as 175 human traits measuring different domains including cognition, substance use, motor, sensory and emotion. The test-retest data validated the developed algorithms. With the HCP data, we found that structural connectomes are associated with a wide range of traits, e.g., fluid intelligence, language comprehension, and motor skills are associated with increased cortical-cortical brain structural connectivity, while the use of alcohol, tobacco, and marijuana are associated with decreased cortical-cortical connectivity. We also demonstrated that our extracted structural connectomes and analysis method can give superior prediction accuracies compared with alternative connectome constructions and other tensor and network regression methods.

Extrema-weighted feature extraction for functional data.

Motivation: Although there is a rich literature on methods for assessing the impact of functional predictors, the focus has been on approaches for dimension reduction that do not suit certain applications. Examples of standard approaches include functional linear models, functional principal components regression and cluster-based approaches, such as latent trajectory analysis. This article is motivated by applications in which the dynamics in a predictor, across times when the value is relatively extreme, are particularly informative about the response. For example, physicians are interested in relating the dynamics of blood pressure changes during surgery to post-surgery adverse outcomes, and it is thought that the dynamics are more important when blood pressure is significantly elevated or lowered. Results: We propose a novel class of extrema-weighted feature (XWF) extraction models. Key components in defining XWFs include the marginal density of the predictor, a function up-weighting values at extreme quantiles of this marginal, and functionals characterizing local dynamics. Algorithms are proposed for fitting of XWF-based regression and classification models, and are compared with current methods for functional predictors in simulations and a blood pressure during surgery application. XWFs find features of intraoperative blood pressure trajectories that are predictive of postoperative mortality. By their nature, most of these features cannot be found by previous methods. Availability and implementation: The R package 'xwf' is available at the CRAN repository: https://cran.r-project.org/package=xwf. Supplementary information: Supplementary data are available at Bioinformatics online.

Enteropathy-associated T cell lymphoma subtypes are characterized by loss of function of SETD2.

Enteropathy-associated T cell lymphoma (EATL) is a lethal, and the most common, neoplastic complication of celiac disease. Here, we defined the genetic landscape of EATL through whole-exome sequencing of 69 EATL tumors. SETD2 was the most frequently silenced gene in EATL (32% of cases). The JAK-STAT pathway was the most frequently mutated pathway, with frequent mutations in STAT5B as well as JAK1, JAK3, STAT3, and SOCS1 We also identified mutations in KRAS, TP53, and TERT Type I EATL and type II EATL (monomorphic epitheliotropic intestinal T cell lymphoma) had highly overlapping genetic alterations indicating shared mechanisms underlying their pathogenesis. We modeled the effects of SETD2 loss in vivo by developing a T cell-specific knockout mouse. These mice manifested an expansion of γδ T cells, indicating novel roles for SETD2 in T cell development and lymphomagenesis. Our data render the most comprehensive genetic portrait yet of this uncommon but lethal disease and may inform future classification schemes.

Bayesian genome- and epigenome-wide association studies with gene level dependence.

High-throughput genetic and epigenetic data are often screened for associations with an observed phenotype. For example, one may wish to test hundreds of thousands of genetic variants, or DNA methylation sites, for an association with disease status. These genomic variables can naturally be grouped by the gene they encode, among other criteria. However, standard practice in such applications is independent screening with a universal correction for multiplicity. We propose a Bayesian approach in which the prior probability of an association for a given genomic variable depends on its gene, and the gene-specific probabilities are modeled nonparametrically. This hierarchical model allows for appropriate gene and genome-wide multiplicity adjustments, and can be incorporated into a variety of Bayesian association screening methodologies with negligible increase in computational complexity. We describe an application to screening for differences in DNA methylation between lower grade glioma and glioblastoma multiforme tumor samples from The Cancer Genome Atlas. Software is available via the package BayesianScreening for R: github.com/lockEF/BayesianScreening.

Shared kernel Bayesian screening.

This article concerns testing for equality of distribution between groups. We focus on screening variables with shared distributional features such as common support, modes and patterns of skewness. We propose a Bayesian testing method using kernel mixtures, which improves performance by borrowing information across the different variables and groups through shared kernels and a common probability of group differences. The inclusion of shared kernels in a finite mixture, with Dirichlet priors on the weights, leads to a simple framework for testing that scales well for high-dimensional data. We provide closed asymptotic forms for the posterior probability of equivalence in two groups and prove consistency under model misspecification. The method is applied to DNA methylation array data from a breast cancer study, and compares favourably to competitors when Type I error is estimated via permutation.

Bayesian Multiscale Modeling of Closed Curves in Point Clouds.

Modeling object boundaries based on image or point cloud data is frequently necessary in medical and scientific applications ranging from detecting tumor contours for targeted radiation therapy, to the classification of organisms based on their structural information. In low-contrast images or sparse and noisy point clouds, there is often insufficient data to recover local segments of the boundary in isolation. Thus, it becomes critical to model the entire boundary in the form of a closed curve. To achieve this, we develop a Bayesian hierarchical model that expresses highly diverse 2D objects in the form of closed curves. The model is based on a novel multiscale deformation process. By relating multiple objects through a hierarchical formulation, we can successfully recover missing boundaries by borrowing structural information from similar objects at the appropriate scale. Furthermore, the model's latent parameters help interpret the population, indicating dimensions of significant structural variability and also specifying a 'central curve' that summarizes the collection. Theoretical properties of our prior are studied in specific cases and efficient Markov chain Monte Carlo methods are developed, evaluated through simulation examples and applied to panorex teeth images for modeling teeth contours and also to a brain tumor contour detection problem.

The genomic landscape of mantle cell lymphoma is related to the epigenetically determined chromatin state of normal B cells.

In this study, we define the genetic landscape of mantle cell lymphoma (MCL) through exome sequencing of 56 cases of MCL. We identified recurrent mutations in ATM, CCND1, MLL2, and TP53. We further identified a number of novel genes recurrently mutated in patients with MCL including RB1, WHSC1, POT1, and SMARCA4. We noted that MCLs have a distinct mutational profile compared with lymphomas from other B-cell stages. The ENCODE project has defined the chromatin structure of many cell types. However, a similar characterization of primary human mature B cells has been lacking. We defined, for the first time, the chromatin structure of primary human naïve, germinal center, and memory B cells through chromatin immunoprecipitation and sequencing for H3K4me1, H3K4me3, H3Ac, H3K36me3, H3K27me3, and PolII. We found that somatic mutations that occur more frequently in either MCLs or Burkitt lymphomas were associated with open chromatin in their respective B cells of origin, naïve B cells, and germinal center B cells. Our work thus elucidates the landscape of gene-coding mutations in MCL and the critical interplay between epigenetic alterations associated with B-cell differentiation and the acquisition of somatic mutations in cancer.

Bayesian consensus clustering.

MOTIVATION: In biomedical research a growing number of platforms and technologies are used to measure diverse but related information, and the task of clustering a set of objects based on multiple sources of data arises in several applications. Most current approaches to multisource clustering either independently determine a separate clustering for each data source or determine a single 'joint' clustering for all data sources. There is a need for more flexible approaches that simultaneously model the dependence and the heterogeneity of the data sources. RESULTS: We propose an integrative statistical model that permits a separate clustering of the objects for each data source. These separate clusterings adhere loosely to an overall consensus clustering, and hence they are not independent. We describe a computationally scalable Bayesian framework for simultaneous estimation of both the consensus clustering and the source-specific clusterings. We demonstrate that this flexible approach is more robust than joint clustering of all data sources, and is more powerful than clustering each data source independently. We present an application to subtype identification of breast cancer tumor samples using publicly available data from The Cancer Genome Atlas. AVAILABILITY: R code with instructions and examples is available at http://people.duke.edu/%7Eel113/software.html.

Genetic heterogeneity of diffuse large B-cell lymphoma.

Diffuse large B-cell lymphoma (DLBCL) is the most common form of lymphoma in adults. The disease exhibits a striking heterogeneity in gene expression profiles and clinical outcomes, but its genetic causes remain to be fully defined. Through whole genome and exome sequencing, we characterized the genetic diversity of DLBCL. In all, we sequenced 73 DLBCL primary tumors (34 with matched normal DNA). Separately, we sequenced the exomes of 21 DLBCL cell lines. We identified 322 DLBCL cancer genes that were recurrently mutated in primary DLBCLs. We identified recurrent mutations implicating a number of known and not previously identified genes and pathways in DLBCL including those related to chromatin modification (ARID1A and MEF2B), NF-κB (CARD11 and TNFAIP3), PI3 kinase (PIK3CD, PIK3R1, and MTOR), B-cell lineage (IRF8, POU2F2, and GNA13), and WNT signaling (WIF1). We also experimentally validated a mutation in PIK3CD, a gene not previously implicated in lymphomas. The patterns of mutation demonstrated a classic long tail distribution with substantial variation of mutated genes from patient to patient and also between published studies. Thus, our study reveals the tremendous genetic heterogeneity that underlies lymphomas and highlights the need for personalized medicine approaches to treating these patients.

The genetic landscape of mutations in Burkitt lymphoma.

Burkitt lymphoma is characterized by deregulation of MYC, but the contribution of other genetic mutations to the disease is largely unknown. Here, we describe the first completely sequenced genome from a Burkitt lymphoma tumor and germline DNA from the same affected individual. We further sequenced the exomes of 59 Burkitt lymphoma tumors and compared them to sequenced exomes from 94 diffuse large B-cell lymphoma (DLBCL) tumors. We identified 70 genes that were recurrently mutated in Burkitt lymphomas, including ID3, GNA13, RET, PIK3R1 and the SWI/SNF genes ARID1A and SMARCA4. Our data implicate a number of genes in cancer for the first time, including CCT6B, SALL3, FTCD and PC. ID3 mutations occurred in 34% of Burkitt lymphomas and not in DLBCLs. We show experimentally that ID3 mutations promote cell cycle progression and proliferation. Our work thus elucidates commonly occurring gene-coding mutations in Burkitt lymphoma and implicates ID3 as a new tumor suppressor gene.

High-dimensional variable selection in meta-analysis for censored data.

This article considers the problem of selecting predictors of time to an event from a high-dimensional set of candidate predictors using data from multiple studies. As an alternative to the current multistage testing approaches, we propose to model the study-to-study heterogeneity explicitly using a hierarchical model to borrow strength. Our method incorporates censored data through an accelerated failure time model. Using a carefully formulated prior specification, we develop a fast approach to predictor selection and shrinkage estimation for high-dimensional predictors. For model fitting, we develop a Monte Carlo expectation maximization (MC-EM) algorithm to accommodate censored data. The proposed approach, which is related to the relevance vector machine (RVM), relies on maximum a posteriori estimation to rapidly obtain a sparse estimate. As for the typical RVM, there is an intrinsic thresholding property in which unimportant predictors tend to have their coefficients shrunk to zero. We compare our method with some commonly used procedures through simulation studies. We also illustrate the method using the gene expression barcode data from three breast cancer studies.

Semiparametric bayes' proportional odds models for current status data with underreporting.

Current status data are a type of interval-censored event time data in which all the individuals are either left or right censored. For example, our motivation is drawn from a cross-sectional study, which measured whether or not fibroid onset had occurred by the age of an ultrasound exam for each woman. We propose a semiparametric Bayesian proportional odds model in which the baseline event time distribution is estimated nonparametrically by using adaptive monotone splines in a logistic regression model and the potential risk factors are included in the parametric part of the mean structure. The proposed approach has the advantage of being straightforward to implement using a simple and efficient Gibbs sampler, whereas alternative semiparametric Bayes' event time models encounter problems for current status data. The model is generalized to allow systematic underreporting in a subset of the data, and the methods are applied to an epidemiologic study of uterine fibroids.

Semiparametric bayes multiple testing: applications to tumor data.

In National Toxicology Program (NTP) studies, investigators want to assess whether a test agent is carcinogenic overall and specific to certain tumor types, while estimating the dose-response profiles. Because there are potentially correlations among the tumors, a joint inference is preferred to separate univariate analyses for each tumor type. In this regard, we propose a random effect logistic model with a matrix of coefficients representing log-odds ratios for the adjacent dose groups for tumors at different sites. We propose appropriate nonparametric priors for these coefficients to characterize the correlations and to allow borrowing of information across different dose groups and tumor types. Global and local hypotheses can be easily evaluated by summarizing the output of a single Monte Carlo Markov chain (MCMC). Two multiple testing procedures are applied for testing local hypotheses based on the posterior probabilities of local alternatives. Simulation studies are conducted and an NTP tumor data set is analyzed illustrating the proposed approach.

Uterine leiomyomata in relation to insulin-like growth factor-I, insulin, and diabetes.

BACKGROUND: Insulin-like growth factor-I (IGF-I) and insulin stimulate cell proliferation in uterine leiomyoma (fibroid) tissue. We hypothesized that circulating levels of these proteins would be associated with increased prevalence and size of uterine fibroids. METHODS: Participants were 35-49-year-old, randomly selected members of an urban health plan who were enrolled in the study in 1996-1999. Premenopausal participants were screened for fibroids with ultrasound. Fasting blood samples were collected. Associations between fibroids and diabetes, plasma IGF-I, IGF binding protein 3 (BP3), and insulin were evaluated for blacks (n = 585) and whites (n = 403) by using multiple logistic regression. RESULTS: IGF-I showed no association with fibroids in blacks, but in whites the adjusted odds ratios (aORs) for both mid and upper tertiles compared with the lowest tertile were 0.6 (95% confidence intervals [CI] = 0.3-1.0 and 0.4-1.1, respectively). Insulin and diabetes both tended to be inversely associated with fibroids in blacks. The insulin association was with large fibroids; aOR for the upper insulin tertile relative to the lowest was 0.4 (0.2-0.9). The aOR for diabetes was 0.5 (0.2-1.0). Associations of insulin and diabetes with fibroids were weak for whites. Binding protein 3 showed no association with fibroids. CONCLUSIONS: Contrary to our hypothesis, high circulating IGF-I and insulin were not related to increased fibroid prevalence. Instead, there was suggestion of the opposite. The inverse association with diabetes, although based on small numbers, is consistent with previously reported findings. Future studies might investigate vascular dysfunction as a mediator between hyperinsulinemia or diabetes and possible reduced risk of fibroids.

Fitting semiparametric random effects models to large data sets.

For large data sets, it can be difficult or impossible to fit models with random effects using standard algorithms due to memory limitations or high computational burdens. In addition, it would be advantageous to use the abundant information to relax assumptions, such as normality of random effects. Motivated by data from an epidemiologic study of childhood growth, we propose a 2-stage method for fitting semiparametric random effects models to longitudinal data with many subjects. In the first stage, we use a multivariate clustering method to identify G<

Bayesian adaptive regression splines for hierarchical data.

This article considers methodology for hierarchical functional data analysis, motivated by studies of reproductive hormone profiles in the menstrual cycle. Current methods standardize the cycle lengths and ignore the timing of ovulation within the cycle, both of which are biologically informative. Methods are needed that avoid standardization, while flexibly incorporating information on covariates and the timing of reference events, such as ovulation and onset of menses. In addition, it is necessary to account for within-woman dependency when data are collected for multiple cycles. We propose an approach based on a hierarchical generalization of Bayesian multivariate adaptive regression splines. Our formulation allows for an unknown set of basis functions characterizing the population-averaged and woman-specific trajectories in relation to covariates. A reversible jump Markov chain Monte Carlo algorithm is developed for posterior computation. Applying the methods to data from the North Carolina Early Pregnancy Study, we investigate differences in urinary progesterone profiles between conception and nonconception cycles.

Association of physical activity with development of uterine leiomyoma.

The relation between physical activity and uterine leiomyomata (fibroids) has received little study, but exercise is protective for breast cancer, another hormonally mediated tumor. Participants in this study were randomly selected members of a health plan based in Washington, DC, aged 35-49 years (734 African Americans, 455 Whites) enrolled between 1996 and 1999. Fibroid status was based on ultrasound screening. Physical activity was based on detailed interview questions. Logistic regression with adjustment for body mass index and other risk factors showed that women in the highest category of physical activity were significantly less likely to have fibroids (odds ratio = 0.6, 95% confidence interval = 0.4, 0.9 for the highest vs. the lowest category (equivalent to approximately > or =7 hours/week vs <2 hours/week)). There was a dose-response pattern; a significant trend was seen for both African-American and White women. A multistate Bayesian analysis indicated that exercise was associated with tumor onset more strongly than with tumor growth. When data for women who reported major fibroid-related symptoms were excluded, results remained essentially unchanged, suggesting that the observed association could not be attributed to reverse causation (fibroids preventing exercise). The authors concluded that regular exercise might help women prevent fibroids.