arXiv subjects

Hongyu Zhao

Publications and source records attributed to Hongyu Zhao.

At least 73 records · Page 4Linked to original sources

arXiv2016

Distance-Correlation based Gene Set Analysis in Longitudinal Studies

Longitudinal gene expression profiles of patients are collected in some clinical studies to monitor disease progression and understand disease etiology. The identification of gene sets that have coordinated changes with relevant clinical outcomes over time from these data could provide significant insights into the molecular basis of disease progression and hence may lead to better treatments. In this article, we propose a Distance-Correlation based Gene Set Analysis (dcGSA) method for longitudinal gene expression data. dcGSA is a non-parametric approach, statistically robust, and can capture both linear and nonlinear relationships between gene sets and clinical outcomes. In addition, dcGSA is able to identify related gene sets in cases where the effects of gene sets on clinical outcomes differ across patients due to the patient heterogeneity, alleviate the confounding effects of some unobserved covariates, and allow the assessment of associations between gene sets and multiple related outcomes simultaneously. Through extensive simulation studies, we demonstrate that dcGSA is more powerful of detecting relevant genes than other commonly used gene set analysis methods. When dcGSA is applied to multiple real datasets of different characteristics, we are able to identify more disease related gene sets than other methods.

stat.ME↗

arXiv2016

A Bayesian Semiparametric Factor Analysis Model for Subtype Identification

Disease subtype identification (clustering) is an important problem in biomedical research. Gene expression profiles are commonly utilized to infer disease subtypes, which often lead to biologically meaningful insights into disease. Despite many successes, existing clustering methods may not perform well when genes are highly correlated and many uninformative genes are included for clustering due to the high dimensionality. In this article, we introduce a novel subtype identification method in the Bayesian setting based on gene expression profiles. This method, called BCSub, adopts an innovative semiparametric Bayesian factor analysis model to reduce the dimension of the data to a few factor scores for clustering. Specifically, the factor scores are assumed to follow the Dirichlet process mixture model in order to induce clustering. Through extensive simulation studies, we show that BCSub has improved performance over commonly used clustering methods. When applied to two gene expression datasets, our model is able to identify subtypes that are clinically more relevant than those identified from the existing methods.

stat.ME↗

arXiv2016

Joint Models for Time-to-Event Data and Longitudinal Biomarkers of High Dimension

Joint models for longitudinal biomarkers and time-to-event data are widely used in longitudinal studies. Many joint modeling approaches have been proposed to deal with different types of longitudinal biomarkers and survival outcomes. However, most existing joint modeling methods cannot deal with a large number of longitudinal biomarkers simultaneously, such as the longitudinally collected gene expression profiles. In this article, we propose a new joint modeling method under the Bayesian framework, which is able to deal with longitudinal biomarkers of high dimension. Specifically, we assume that only a few unobserved latent variables are related to the survival outcome and the latent variables are inferred using a factor analysis model, which greatly reduces the dimensionality of the biomarkers and also accounts for the high correlations among the biomarkers. Through extensive simulation studies, we show that our proposed method has improved prediction accuracy over other joint modeling methods. We illustrate the usefulness of our method on a dataset of idiopathic pulmonary fibrosis patients in which we are interested in predicting the patients' time-to-death using their gene expression profiles.

stat.ME↗

arXiv2016

Ultrahigh Dimensional Feature Selection via Kernel Canonical Correlation Analysis

High-dimensional variable selection is an important issue in many scientific fields, such as genomics. In this paper, we develop a sure independence feature screening pro- cedure based on kernel canonical correlation analysis (KCCA-SIS, for short). KCCA- SIS is easy to be implemented and applied. Compared to the sure independence screen- ing procedure based on the Pearson correlation (SIS, for short) developed by Fan and Lv [2008], KCCA-SIS can handle nonlinear dependencies among variables. Compared to the sure independence screening procedure based on the distance correlation (DC- SIS, for short) proposed by Li et al. [2012], KCCA-SIS is scale free, distribution free and has better approximation results based on the universal characteristic of Gaussian Kernel (Micchelli et al. [2006]). KCCA-SIS is more general than SIS and DC-SIS in the sense that SIS and DC-SIS correspond to certain choice of kernels. Compared to supremum of Hilbert Schmidt independence criterion-Sure independence screening (sup-HSIC-SIS, for short) developed by Balasubramanian et al. [2013], KCCA-SIS is scale free removing the marginal variation of features and response variables. No model assumption is needed between response and predictors to apply KCCA-SIS and it can be used in ultrahigh dimensional data analysis. Similar to DC-SIS and sup- HSIC-SIS, KCCA-SIS can also be used directly to screen grouped predictors and for multivariate response variables. We show that KCCA-SIS has the sure screening prop- erty, and has better performance through simulation studies. We applied KCCA-SIS to study Autism genes in a spatiotemporal gene expression dataset for human brain development, and obtained better results based on gene ontology enrichment analysis comparing to the other existing methods.

stat.ME↗

arXiv2015

On Joint Estimation of Gaussian Graphical Models for Spatial and Temporal Data

In this paper, we first propose a Bayesian neighborhood selection method to estimate Gaussian Graphical Models (GGMs). We show the graph selection consistency of this method in the sense that the posterior probability of the true model converges to one. When there are multiple groups of data available, instead of estimating the networks independently for each group, joint estimation of the networks may utilize the shared information among groups and lead to improved estimation for each individual network. Our method is extended to jointly estimate GGMs in multiple groups of data with complex structures, including spatial data, temporal data and data with both spatial and temporal structures. Markov random field (MRF) models are used to efficiently incorporate the complex data structures. We develop and implement an efficient algorithm for statistical inference that enables parallel computing. Simulation studies suggest that our approach achieves better accuracy in network estimation compared with methods not incorporating spatial and temporal dependencies when there are shared structures among the networks, and that it performs comparably well otherwise. Finally, we illustrate our method using the human brain gene expression microarray dataset, where the expression levels of genes are measured in different brain regions across multiple time periods.

stat.AP↗

arXiv2015

A Markov random field-based approach to characterizing human brain development using spatial-temporal transcriptome data

Human neurodevelopment is a highly regulated biological process. In this article, we study the dynamic changes of neurodevelopment through the analysis of human brain microarray data, sampled from 16 brain regions in 15 time periods of neurodevelopment. We develop a two-step inferential procedure to identify expressed and unexpressed genes and to detect differentially expressed genes between adjacent time periods. Markov Random Field (MRF) models are used to efficiently utilize the information embedded in brain region similarity and temporal dependency in our approach. We develop and implement a Monte Carlo expectation-maximization (MCEM) algorithm to estimate the model parameters. Simulation studies suggest that our approach achieves lower misclassification error and potential gain in power compared with models not incorporating spatial similarity and temporal dependency.

stat.AP↗

arXiv2015

Posterior Contraction Rates of the Phylogenetic Indian Buffet Processes

By expressing prior distributions as general stochastic processes, nonparametric Bayesian methods provide a flexible way to incorporate prior knowledge and constrain the latent structure in statistical inference. The Indian buffet process (IBP) is such an example that can be used to define a prior distribution on infinite binary features, where the exchangeability among subjects is assumed. The phylogenetic Indian buffet process (pIBP), a derivative of IBP, enables the modeling of non-exchangeability among subjects through a stochastic process on a rooted tree, which is similar to that used in phylogenetics, to describe relationships among the subjects. In this paper, we study the theoretical properties of IBP and pIBP under a binary factor model. We establish the posterior contraction rates for both IBP and pIBP and substantiate the theoretical results through simulation studies. This is the first work addressing the frequentist property of the posterior behaviors of IBP and pIBP. We also demonstrated its practical usefulness by applying pIBP prior to a real data example arising in the field of cancer genomics where the exchangeability among subjects is violated.

stat.ML↗

arXiv2015

NBLDA: Negative Binomial Linear Discriminant Analysis for RNA-Seq Data

RNA-sequencing (RNA-Seq) has become a powerful technology to characterize gene expression profiles because it is more accurate and comprehensive than microarrays. Although statistical methods that have been developed for microarray data can be applied to RNA-Seq data, they are not ideal due to the discrete nature of RNA-Seq data. The Poisson distribution and negative binomial distribution are commonly used to model count data. Recently, Witten (2011) proposed a Poisson linear discriminant analysis for RNA-Seq data. The Poisson assumption may not be as appropriate as negative binomial distribution when biological replicates are available and in the presence of overdispersion (i.e., when the variance is larger than the mean). However, it is more complicated to model negative binomial variables because they involve a dispersion parameter that needs to be estimated. In this paper, we propose a negative binomial linear discriminant analysis for RNA-Seq data. By Bayes' rule, we construct the classifier by fitting a negative binomial model, and propose some plug-in rules to estimate the unknown parameters in the classifier. The relationship between the negative binomial classifier and the Poisson classifier is explored, with a numerical investigation of the impact of dispersion on the discriminant score. Simulation results show the superiority of our proposed method. We also analyze four real RNA-Seq data sets to demonstrate the advantage of our method in real-world applications.

stat.AP↗

arXiv2014

Low-Rank Modeling and Its Applications in Image Analysis

Low-rank modeling generally refers to a class of methods that solve problems by representing variables of interest as low-rank matrices. It has achieved great success in various fields including computer vision, data mining, signal processing and bioinformatics. Recently, much progress has been made in theories, algorithms and applications of low-rank modeling, such as exact low-rank matrix recovery via convex programming and matrix completion applied to collaborative filtering. These advances have brought more and more attentions to this topic. In this paper, we review the recent advance of low-rank modeling, the state-of-the-art algorithms, and related applications in image analysis. We first give an overview to the concept of low-rank modeling and challenging problems in this area. Then, we summarize the models and algorithms for low-rank matrix recovery and illustrate their advantages and limitations with numerical experiments. Next, we introduce a few applications of low-rank modeling in the context of image analysis. Finally, we conclude this paper with some discussions.

cs.CV↗

arXiv2014

Estimation for ultra-high dimensional factor model: a pivotal variable detection based approach

For factor model, the involved covariance matrix often has no row sparse structure because the common factors may lead some variables to strongly associate with many others. Under the ultra-high dimensional paradigm, this feature causes existing methods for sparse covariance matrix in the literature not directly applicable. In this paper, for general covariance matrix, a novel approach to detect these variables that is called the pivotal variables is suggested. Then, two-stage estimation procedures are proposed to handle ultra-high dimensionality in factor model. In these procedures, pivotal variable detection is performed as a screening step and then existing approaches are applied to refine the working model. The estimation efficiency can be promoted under weaker assumptions on the model structure. Simulations are conducted to examine the performance of the new method and a real dataset is analysed for illustration.

stat.ME↗

arXiv2014

Detection boundary and Higher Criticism approach for rare and weak genetic effects

Genome-wide association studies (GWAS) have identified many genetic factors underlying complex human traits. However, these factors have explained only a small fraction of these traits' genetic heritability. It is argued that many more genetic factors remain undiscovered. These genetic factors likely are weakly associated at the population level and sparsely distributed across the genome. In this paper, we adapt the recent innovations on Tukey's Higher Criticism (Tukey [The Higher Criticism (1976) Princeton Univ.]; Donoho and Jin [Ann. Statist. 32 (2004) 962-994]) to SNP-set analysis of GWAS, and develop a new theoretical framework in large-scale inference to assess the joint significance of such rare and weak effects for a quantitative trait. In the core of our theory is the so-called detection boundary, a curve in the two-dimensional phase space that quantifies the rarity and strength of genetic effects. Above the detection boundary, the overall effects of genetic factors are strong enough for reliable detection. Below the detection boundary, the genetic factors are simply too rare and too weak for reliable detection. We show that the HC-type methods are optimal in that they reliably yield detection once the parameters of the genetic effects fall above the detection boundary and that many commonly used SNP-set methods are suboptimal. The superior performance of the HC-type approach is demonstrated through simulations and the analysis of a GWAS data set of Crohn's disease.

stat.AP↗

arXiv2014

Change Point Analysis of Histone Modifications Reveals Epigenetic Blocks Linking to Physical Domains

Histone modification is a vital epigenetic mechanism for transcriptional control in eukaryotes. High-throughput techniques have enabled whole-genome analysis of histone modifications in recent years. However, most studies assume one combination of histone modification invariantly translates to one transcriptional output regardless of local chromatin environment. In this study we hypothesize that, the genome is organized into local domains that manifest similar enrichment pattern of histone modification, which leads to orchestrated regulation of expression of genes with relevant bio- logical functions. We propose a multivariate Bayesian Change Point (BCP) model to segment the Drosophila melanogaster genome into consecutive blocks on the basis of combinatorial patterns of histone marks. By modeling the sparse distribution of histone marks across the chromosome with a zero-inflated Gaussian mixture, our partitions capture local BLOCKs that manifest relatively homogeneous enrichment pattern of histone modifications. We further characterized BLOCKs by their transcription levels, distribution of genes, degree of co-regulation and GO enrichment. Our results demonstrate that these BLOCKs, although inferred merely from histone modifications, reveal strong relevance with physical domains, which suggest their important roles in chromatin organization and coordinated gene regulation.

q-bio.GN↗

arXiv2014

High-dimensional genome-wide association study and misspecified mixed model analysis

We study behavior of the restricted maximum likelihood (REML) estimator under a misspecified linear mixed model (LMM) that has received much attention in recent gnome-wide association studies. The asymptotic analysis establishes consistency of the REML estimator of the variance of the errors in the LMM, and convergence in probability of the REML estimator of the variance of the random effects in the LMM to a certain limit, which is equal to the true variance of the random effects multiplied by the limiting proportion of the nonzero random effects present in the LMM. The aymptotic results also establish convergence rate (in probability) of the REML estimators as well as a result regarding convergence of the asymptotic conditional variance of the REML estimator. The asymptotic results are fully supported by the results of empirical studies, which include extensive simulation studies that compare the performance of the REML estimator (under the misspecified LMM) with other existing methods.

math.ST↗

arXiv2014

GPA: A statistical approach to prioritizing GWAS results by integrating pleiotropy information and annotation data

Genome-wide association studies (GWAS) suggests that a complex disease is typically affected by many genetic variants with small or moderate effects. Identification of these risk variants remains to be a very challenging problem. Traditional approaches focusing on a single GWAS dataset alone ignore relevant information that could potentially improve our ability to detect these variants. We proposed a novel statistical approach, named GPA, to performing integrative analysis of multiple GWAS datasets and functional annotations. Hypothesis testing procedures were developed to facilitate statistical inference of pleiotropy and enrichment of functional annotation. We applied our approach to perform systematic analysis of five psychiatric disorders. Not only did GPA identify many weak signals missed by the original single phenotype analysis, but also revealed interesting genetic architectures of these disorders. We also applied GPA to the bladder cancer GWAS data with the ENCODE DNase-seq data from 125 cell lines and showed that GPA can detect cell lines that are more biologically relevant to the phenotype of interest.

stat.ME↗

arXiv2013

A Spatial Simulation Approach to Account for Protein Structure When Identifying Non-Random Somatic Mutations

Background: Current research suggests that a small set of "driver" mutations are responsible for tumorigenesis while a larger body of "passenger" mutations occurs in the tumor but does not progress the disease. Due to recent pharmacological successes in treating cancers caused by driver mutations, a variety of of methodologies that attempt to identify such mutations have been developed. Based on the hypothesis that driver mutations tend to cluster in key regions of the protein, the development of cluster identification algorithms has become critical. Results: We have developed a novel methodology, SpacePAC (Spatial Protein Amino acid Clustering), that identifies mutational clustering by considering the protein tertiary structure directly in 3D space. By combining the mutational data in the Catalogue of Somatic Mutations in Cancer (COSMIC) and the spatial information in the Protein Data Bank (PDB), SpacePAC is able to identify novel mutation clusters in many proteins such as FGFR3 and CHRM2. In addition, SpacePAC is better able to localize the most significant mutational hotspots as demonstrated in the cases of BRAF and ALK. The R package is available on Bioconductor at: http://www.bioconductor.org/packages/release/bioc/html/SpacePAC.html Conclusion: SpacePAC adds a valuable tool to the identification of mutational clusters while considering protein tertiary structure

q-bio.GN↗

arXiv2013

Asymptotically Normal and Efficient Estimation of Covariate-Adjusted Gaussian Graphical Model

A tuning-free procedure is proposed to estimate the covariate-adjusted Gaussian graphical model. For each finite subgraph, this estimator is asymptotically normal and efficient. As a consequence, a confidence interval can be obtained for each edge. The procedure enjoys easy implementation and efficient computation through parallel estimation on subgraphs or edges. We further apply the asymptotic normality result to perform support recovery through edge-wise adaptive thresholding. This support recovery procedure is called ANTAC, standing for Asymptotically Normal estimation with Thresholding after Adjusting Covariates. ANTAC outperforms other methodologies in the literature in a range of simulation studies. We apply ANTAC to identify gene-gene interactions using an eQTL dataset. Our result achieves better interpretability and accuracy in comparison with CAMPE.

stat.ME↗

arXiv2013

Improving genetic risk prediction by leveraging pleiotropy

An important task of human genetics studies is to accurately predict disease risks in individuals based on genetic markers, which allows for identifying individuals at high disease risks, and facilitating their disease treatment and prevention. Although hundreds of genome-wide association studies (GWAS) have been conducted on many complex human traits in recent years, there has been only limited success in translating these GWAS data into clinically useful risk prediction models. The predictive capability of GWAS data is largely bottlenecked by the available training sample size due to the presence of numerous variants carrying only small to modest effects. Recent studies have shown that different human traits may share common genetic bases. Therefore, an attractive strategy to increase the training sample size and hence improve the prediction accuracy is to integrate data of genetically correlated phenotypes. Yet the utility of genetic correlation in risk prediction has not been explored in the literature. In this paper, we analyzed GWAS data for bipolar and related disorders (BARD) and schizophrenia (SZ) with a bivariate ridge regression method, and found that jointly predicting the two phenotypes could substantially increase prediction accuracy as measured by the AUC (area under the receiver operating characteristic curve). We also found similar prediction accuracy improvements when we jointly analyzed GWAS data for Crohn's disease (CD) and ulcerative colitis (UC). The empirical observations were substantiated through our comprehensive simulation studies, suggesting that a gain in prediction accuracy can be obtained by combining phenotypes with relatively high genetic correlations. Through both real data and simulation studies, we demonstrated pleiotropy as a valuable asset that opens up a new opportunity to improve genetic risk prediction in the future.

q-bio.GN↗

arXiv2013

A Graph Theoretic Approach to Utilizing Protein Structure to Identify Non-Random Somatic Mutations

Background: It is well known that the development of cancer is caused by the accumulation of somatic mutations within the genome. For oncogenes specifically, current research suggests that there is a small set of "driver" mutations that are primarily responsible for tumorigenesis. Further, due to some recent pharmacological successes in treating these driver mutations and their resulting tumors, a variety of methods have been developed to identify potential driver mutations using methods such as machine learning and mutational clustering. We propose a novel methodology that increases our power to identify mutational clusters by taking into account protein tertiary structure via a graph theoretical approach. Results: We have designed and implemented GraphPAC (Graph Protein Amino Acid Clustering) to identify mutational clustering while considering protein spatial structure. Using GraphPAC, we are able to detect novel clusters in proteins that are known to exhibit mutation clustering as well as identify clusters in proteins without evidence of prior clustering based on current methods. Specifically, by utilizing the spatial information available in the Protein Data Bank (PDB) along with the mutational data in the Catalogue of Somatic Mutations in Cancer (COSMIC), GraphPAC identifies new mutational clusters in well known oncogenes such as EGFR and KRAS. Further, by utilizing graph theory to account for the tertiary structure, GraphPAC identifies clusters in DPP4, NRP1 and other proteins not identified by existing methods. The R package is available at: http://bioconductor.org/packages/release/bioc/html/GraphPAC.html Conclusion: GraphPAC provides an alternative to iPAC and an extension to current methodology when identifying potential activating driver mutations by utilizing a graph theoretic approach when considering protein tertiary structure.

q-bio.GN↗