Jianhua Hu
Current Organizations / Institutes
The organizations/institutes from the scholar's latest article
Previous Organizations / Institutes
The organizations / institutes from the scholar's other articles in the past ten years
Research Subjects
Jianhua Hu has a total of 33 co-authors. Here are the top co-authors by number of shared publications:
- Xuming He
Department of Statistics and Data Science, Washington University in St. Louis
3 shared publications - Yoonsuh Jung
Department of Statistics, Korea University
3 shared publications Xuming He
Department of Statistics, University of Illinois at Urbana-Champaign, Champaign, IL, 61820, USA, 2Department of Biostatistics, University of Texas M.D. Anderson Cancer Center, Houston, TX, 77230, USA and 3Department of Statistics, University of Michigan, Ann Arbor, MI, 48109, USA
2 shared publications- Adarsh Joshi1 shared publication
- Annie Qu
Department of Statistics and Applied Probability, University of California, Santa Barbara, CA, USA
1 shared publication Bin Bin Li
Department of Statistics, University of Illinois at Urbana-Champaign, Champaign, IL, 61820, USA, 2Department of Biostatistics, University of Texas M.D. Anderson Cancer Center, Houston, TX, 77230, USA and 3Department of Statistics, University of Michigan, Ann Arbor, MI, 48109, USA
1 shared publication
Latest Publications
- 31 January 2018
- Published by SAGE Publications in Statistical Methods in Medical Research
- Vol. 28 (4) , 1230-1246
Transformed low-rank ANOVA models for high-dimensional variable selection
High-dimensional data are often encountered in biomedical, environmental, and other studies. For example, in biomedical studies that involve high-throughput omic data, an important problem is to search for genetic variables that are predictive of a particular phenotype. A conventional solution is to characterize such relationships through regression models in which a phenotype is treated as the response variable and the variables are treated as covariates; this approach becomes particularly challenging when the number of variables exceeds the number of samples. We propose a general framework for expressing the transformed mean of high-dimensional variables in an exponential distribution family via ANOVA models in which a low-rank interaction space captures the association between the phenotype and the variables. This alternative method transforms the variable selection problem into a well-posed problem with the number of observations larger than the number of variables. In addition, we propose a model selection criterion for the new model framework with a diverging number of parameters, and establish the consistency of the selection criterion. We demonstrate the appealing performance of the proposed method in terms of prediction and detection accuracy through simulations and real data analyses.
- 31 January 2018
- Published by SAGE Publications in Statistical Methods in Medical Research
- Vol. 28 (4) , 1230-1246
Transformed low-rank ANOVA models for high-dimensional variable selection
High-dimensional data are often encountered in biomedical, environmental, and other studies. For example, in biomedical studies that involve high-throughput omic data, an important problem is to search for genetic variables that are predictive of a particular phenotype. A conventional solution is to characterize such relationships through regression models in which a phenotype is treated as the response variable and the variables are treated as covariates; this approach becomes particularly challenging when the number of variables exceeds the number of samples. We propose a general framework for expressing the transformed mean of high-dimensional variables in an exponential distribution family via ANOVA models in which a low-rank interaction space captures the association between the phenotype and the variables. This alternative method transforms the variable selection problem into a well-posed problem with the number of observations larger than the number of variables. In addition, we propose a model selection criterion for the new model framework with a diverging number of parameters, and establish the consistency of the selection criterion. We demonstrate the appealing performance of the proposed method in terms of prediction and detection accuracy through simulations and real data analyses.
- Open Access
- 5 July 2016
- Published by Oxford University Press (OUP) in Bioinformatics
- Vol. 32 (21) , 3279-3290
SIFORM: shared informative factor models for integration of multi-platform bioinformatic data
Motivation: High-dimensional omic data derived from different technological platforms have been extensively used to facilitate comprehensive understanding of disease mechanisms and to determine personalized health treatments. Numerous studies have integrated multi-platform omic data; however, few have efficiently and simultaneously addressed the problems that arise from high dimensionality and complex correlations. Results: We propose a statistical framework of shared informative factor models that can jointly analyze multi-platform omic data and explore their associations with a disease phenotype. The common disease-associated sample characteristics across different data types can be captured through the shared structure space, while the corresponding weights of genetic variables directly index the strengths of their association with the phenotype. Extensive simulation studies demonstrate the performance of the proposed method in terms of biomarker detection accuracy via comparisons with three popular regularized regression methods. We also apply the proposed method to The Cancer Genome Atlas lung adenocarcinoma dataset to jointly explore associations of mRNA expression and protein expression with smoking status. Many of the identified biomarkers belong to key pathways for lung tumorigenesis, some of which are known to show differential expression across smoking levels. We discover potential biomarkers that reveal different mechanisms of lung tumorigenesis between light smokers and heavy smokers. Availability and Implementation: R code to implement the new method can be downloaded from http://odin.mdacc.tmc.edu/jhhu/ Contact:jhu@mdanderson.org
- 5 July 2016
- Published by Oxford University Press (OUP) in Bioinformatics
- Vol. 32 (21) , 3279-3290
Open AccessSIFORM: shared informative factor models for integration of multi-platform bioinformatic data
Motivation: High-dimensional omic data derived from different technological platforms have been extensively used to facilitate comprehensive understanding of disease mechanisms and to determine personalized health treatments. Numerous studies have integrated multi-platform omic data; however, few have efficiently and simultaneously addressed the problems that arise from high dimensionality and complex correlations. Results: We propose a statistical framework of shared informative factor models that can jointly analyze multi-platform omic data and explore their associations with a disease phenotype. The common disease-associated sample characteristics across different data types can be captured through the shared structure space, while the corresponding weights of genetic variables directly index the strengths of their association with the phenotype. Extensive simulation studies demonstrate the performance of the proposed method in terms of biomarker detection accuracy via comparisons with three popular regularized regression methods. We also apply the proposed method to The Cancer Genome Atlas lung adenocarcinoma dataset to jointly explore associations of mRNA expression and protein expression with smoking status. Many of the identified biomarkers belong to key pathways for lung tumorigenesis, some of which are known to show differential expression across smoking levels. We discover potential biomarkers that reveal different mechanisms of lung tumorigenesis between light smokers and heavy smokers. Availability and Implementation: R code to implement the new method can be downloaded from http://odin.mdacc.tmc.edu/jhhu/ Contact:jhu@mdanderson.org
- 8 March 2016
- Published by Oxford University Press (OUP) in Biometrics
- Vol. 72 (3) , 815-826
Sequential Model Selection-Based Segmentation to Detect DNA Copy Number Variation
Summary: Array-based CGH experiments are designed to detect genomic aberrations or regions of DNA copy-number variation that are associated with an outcome, typically a state of disease. Most of the existing statistical methods target on detecting DNA copy number variations in a single sample or array. We focus on the detection of group effect variation, through simultaneous study of multiple samples from multiple groups. Rather than using direct segmentation or smoothing techniques, as commonly seen in existing detection methods, we develop a sequential model selection procedure that is guided by a modified Bayesian information criterion. This approach improves detection accuracy by accumulatively utilizing information across contiguous clones, and has computational advantage over the existing popular detection methods. Our empirical investigation suggests that the performance of the proposed method is superior to that of the existing detection methods, in particular, in detecting small segments or separating neighboring segments with differential degrees of copy-number variation.
- 8 March 2016
- Published by Oxford University Press (OUP) in Biometrics
- Vol. 72 (3) , 815-826
Sequential Model Selection-Based Segmentation to Detect DNA Copy Number Variation
Summary: Array-based CGH experiments are designed to detect genomic aberrations or regions of DNA copy-number variation that are associated with an outcome, typically a state of disease. Most of the existing statistical methods target on detecting DNA copy number variations in a single sample or array. We focus on the detection of group effect variation, through simultaneous study of multiple samples from multiple groups. Rather than using direct segmentation or smoothing techniques, as commonly seen in existing detection methods, we develop a sequential model selection procedure that is guided by a modified Bayesian information criterion. This approach improves detection accuracy by accumulatively utilizing information across contiguous clones, and has computational advantage over the existing popular detection methods. Our empirical investigation suggests that the performance of the proposed method is superior to that of the existing detection methods, in particular, in detecting small segments or separating neighboring segments with differential degrees of copy-number variation.
- 3 April 2015
- Published by Taylor & Francis in Journal of Computational and Graphical Statistics
- Vol. 24 (2) , 455-476
Estimating and Identifying Unspecified Correlation Structure for Longitudinal Data
Identifying correlation structure is important to achieving estimation efficiency in analyzing longitudinal data, and is also crucial for drawing valid statistical inference for large size clustered data. In this paper, we propose a nonparametric method to estimate the correlation structure, which is applicable for discrete longitudinal data. We utilize eigenvector-based basis matrices to approximate the inverse of the empirical correlation matrix and determine the number of basis matrices via model selection. A penalized objective function based on the difference between the empirical and model approximation of the correlation matrices is adopted to select an informative structure for the correlation matrix. The eigenvector representation of the correlation estimation is capable of reducing the risk of model misspecification, and also provides useful information on the specific within-cluster correlation pattern of the data. We show that the proposed method possesses the oracle property and selects the true correlation structure consistently. The proposed method is illustrated through simulations and two data examples on air pollution and sonar signal studies.
- 3 April 2015
- Published by Taylor & Francis in Journal of Computational and Graphical Statistics
- Vol. 24 (2) , 455-476
Estimating and Identifying Unspecified Correlation Structure for Longitudinal Data
Identifying correlation structure is important to achieving estimation efficiency in analyzing longitudinal data, and is also crucial for drawing valid statistical inference for large size clustered data. In this paper, we propose a nonparametric method to estimate the correlation structure, which is applicable for discrete longitudinal data. We utilize eigenvector-based basis matrices to approximate the inverse of the empirical correlation matrix and determine the number of basis matrices via model selection. A penalized objective function based on the difference between the empirical and model approximation of the correlation matrices is adopted to select an informative structure for the correlation matrix. The eigenvector representation of the correlation estimation is capable of reducing the risk of model misspecification, and also provides useful information on the specific within-cluster correlation pattern of the data. We show that the proposed method possesses the oracle property and selects the true correlation structure consistently. The proposed method is illustrated through simulations and two data examples on air pollution and sonar signal studies.
- 6 February 2015
- Published by Taylor & Francis in Journal of Nonparametric Statistics
- Vol. 27 (2) , 167-179
AK-fold averaging cross-validation procedure
Cross-validation (CV) type of methods have been widely used to facilitate model estimation and variable selection. In this work, we suggest a new K-fold CV procedure to select a candidate ‘optimal’ model from each hold-out fold and average the K candidate ‘optimal’ models to obtain the ultimate model. Due to the averaging effect, the variance of the proposed estimates can be significantly reduced. This new procedure results in more stable and efficient parameter estimation than the classical K-fold CV procedure. In addition, we show the asymptotic equivalence between the proposed and classical CV procedures in the linear regression setting. We also demonstrate the broad applicability of the proposed procedure via two examples of parameter sparsity regularisation and quantile smoothing splines modelling. We illustrate the promise of the proposed method through simulations and a real data example.
- 6 February 2015
- Published by Taylor & Francis in Journal of Nonparametric Statistics
- Vol. 27 (2) , 167-179
AK-fold averaging cross-validation procedure
Cross-validation (CV) type of methods have been widely used to facilitate model estimation and variable selection. In this work, we suggest a new K-fold CV procedure to select a candidate ‘optimal’ model from each hold-out fold and average the K candidate ‘optimal’ models to obtain the ultimate model. Due to the averaging effect, the variance of the proposed estimates can be significantly reduced. This new procedure results in more stable and efficient parameter estimation than the classical K-fold CV procedure. In addition, we show the asymptotic equivalence between the proposed and classical CV procedures in the linear regression setting. We also demonstrate the broad applicability of the proposed procedure via two examples of parameter sparsity regularisation and quantile smoothing splines modelling. We illustrate the promise of the proposed method through simulations and a real data example.
- 2 October 2014
- Published by Taylor & Francis in Journal of the American Statistical Association
- Vol. 109 (508) , 1355-1367
Biomarker Detection in Association Studies: Modeling SNPs Simultaneously via Logistic ANOVA
In genome-wide association studies, the primary task is to detect biomarkers in the form of single nucleotide polymorphisms (SNPs) that have nontrivial associations with a disease phenotype and some other important clinical/environmental factors. However, the extremely large number of SNPs compared to the sample size inhibits application of classical methods such as the multiple logistic regression. Currently, the most commonly used approach is still to analyze one SNP at a time. In this article, we propose to consider the genotypes of the SNPs simultaneously via a logistic analysis of variance (ANOVA) model, which expresses the logit transformed mean of SNP genotypes as the summation of the SNP effects, effects of the disease phenotype and/or other clinical variables, and the interaction effects. We use a reduced-rank representation of the interaction-effect matrix for dimensionality reduction, and employ the L1-penalty in a penalized likelihood framework to filter out the SNPs that have no associations. We develop a majorization–minimization algorithm for computational implementation. In addition, we propose a modified BIC criterion to select the penalty parameters and determine the rank number. The proposed method is applied to a multiple sclerosis dataset and simulated datasets and shows promise in biomarker detection.
- 2 October 2014
- Published by Taylor & Francis in Journal of the American Statistical Association
- Vol. 109 (508) , 1355-1367
Biomarker Detection in Association Studies: Modeling SNPs Simultaneously via Logistic ANOVA
In genome-wide association studies, the primary task is to detect biomarkers in the form of single nucleotide polymorphisms (SNPs) that have nontrivial associations with a disease phenotype and some other important clinical/environmental factors. However, the extremely large number of SNPs compared to the sample size inhibits application of classical methods such as the multiple logistic regression. Currently, the most commonly used approach is still to analyze one SNP at a time. In this article, we propose to consider the genotypes of the SNPs simultaneously via a logistic analysis of variance (ANOVA) model, which expresses the logit transformed mean of SNP genotypes as the summation of the SNP effects, effects of the disease phenotype and/or other clinical variables, and the interaction effects. We use a reduced-rank representation of the interaction-effect matrix for dimensionality reduction, and employ the L1-penalty in a penalized likelihood framework to filter out the SNPs that have no associations. We develop a majorization–minimization algorithm for computational implementation. In addition, we propose a modified BIC criterion to select the penalty parameters and determine the rank number. The proposed method is applied to a multiple sclerosis dataset and simulated datasets and shows promise in biomarker detection.
- Open Access
- 4 March 2014
- Published by Springer Nature in BMC Cancer
- Vol. 14 (1) , 147
Invasive neuroendocrine carcinoma of the breast: a population-based study from the surveillance, epidemiology and end results (SEER) database
Background: Neuroendocrine carcinoma (NEC) of the breast is a rare type of carcinoma that has not been well studied or characterized. Of the limited number of studies reported in the literature, most are case reports. A few small retrospective series studies have been reported. Methods: We reviewed data on 142 cases of mammary NEC recorded in the surveillance, epidemiology, and end results (SEER) database during 2003–2009 and evaluated disease incidence and patient age, sex, and race/ethnicity; clinicopathologic characteristics; and survival in comparison to invasive mammary carcinoma, not otherwise specified. We also performed univariate and multivariate analyses to identify prognostic factors in this disease. Results: Review of the 142 SEER cases revealed that NEC is an aggressive variant of invasive mammary carcinoma. It generally occurred in older women (>60 years); present with larger tumor size (>20 mm), higher histologic grade, and higher clinical stage; and result in shorter overall survival and disease-specific survival than invasive mammary carcinoma, not otherwise specified (IMC-NOS). Overall survival and disease-specific survival were shorter in NEC at each stage than in IMC-NOS of the same stage. Furthermore, when all NEC and IMC-NOS cases were pooled together, neuroendocrine differentiation itself was an adverse prognostic factor independent of other known prognostic factors, including age, tumor size, nodal status, histologic grade, estrogen/progesterone receptor status, and therapy. Conclusions: NEC is a rare but aggressive type of mammary carcinoma. Novel therapeutic approaches should be explored for this uniquely clinical entity.
- 4 March 2014
- Published by Springer Nature in BMC Cancer
- Vol. 14 (1) , 147
Open AccessInvasive neuroendocrine carcinoma of the breast: a population-based study from the surveillance, epidemiology and end results (SEER) database
Background: Neuroendocrine carcinoma (NEC) of the breast is a rare type of carcinoma that has not been well studied or characterized. Of the limited number of studies reported in the literature, most are case reports. A few small retrospective series studies have been reported. Methods: We reviewed data on 142 cases of mammary NEC recorded in the surveillance, epidemiology, and end results (SEER) database during 2003–2009 and evaluated disease incidence and patient age, sex, and race/ethnicity; clinicopathologic characteristics; and survival in comparison to invasive mammary carcinoma, not otherwise specified. We also performed univariate and multivariate analyses to identify prognostic factors in this disease. Results: Review of the 142 SEER cases revealed that NEC is an aggressive variant of invasive mammary carcinoma. It generally occurred in older women (>60 years); present with larger tumor size (>20 mm), higher histologic grade, and higher clinical stage; and result in shorter overall survival and disease-specific survival than invasive mammary carcinoma, not otherwise specified (IMC-NOS). Overall survival and disease-specific survival were shorter in NEC at each stage than in IMC-NOS of the same stage. Furthermore, when all NEC and IMC-NOS cases were pooled together, neuroendocrine differentiation itself was an adverse prognostic factor independent of other known prognostic factors, including age, tumor size, nodal status, histologic grade, estrogen/progesterone receptor status, and therapy. Conclusions: NEC is a rare but aggressive type of mammary carcinoma. Novel therapeutic approaches should be explored for this uniquely clinical entity.
- 26 July 2012
- Published by Oxford University Press (OUP) in Biometrics
- Vol. 68 (3) , 784-792
Analyzing Multiple‐Probe Microarray: Estimation and Application of Gene Expression Indexes
Summary Gene expression index estimation is an essential step in analyzing multiple probe microarray data. Various modeling methods have been proposed in this area. Amidst all, a popular method proposed in Li and Wong (2001) is based on a multiplicative model, which is similar to the additive model discussed in Irizarry et al. (2003a) at the logarithm scale. Along this line, Hu et al. (2006) proposed data transformation to improve expression index estimation based on an ad hoc entropy criteria and naive grid search approach. In this work, we re-examined this problem using a new profile likelihood-based transformation estimation approach that is more statistically elegant and computationally efficient. We demonstrate the applicability of the proposed method using a benchmark Affymetrix U95A spiked-in experiment. Moreover, We introduced a new multivariate expression index and used the empirical study to shows its promise in terms of improving model fitting and power of detecting differential expression over the commonly used univariate expression index. As the other important content of the work, we discussed two generally encountered practical issues in application of gene expression index: normalization and summary statistic used for detecting differential expression. Our empirical study shows somewhat different findings from the MAQC project ( MAQC, 2006 ).
- 26 July 2012
- Published by Oxford University Press (OUP) in Biometrics
- Vol. 68 (3) , 784-792
Analyzing Multiple‐Probe Microarray: Estimation and Application of Gene Expression Indexes
Summary Gene expression index estimation is an essential step in analyzing multiple probe microarray data. Various modeling methods have been proposed in this area. Amidst all, a popular method proposed in Li and Wong (2001) is based on a multiplicative model, which is similar to the additive model discussed in Irizarry et al. (2003a) at the logarithm scale. Along this line, Hu et al. (2006) proposed data transformation to improve expression index estimation based on an ad hoc entropy criteria and naive grid search approach. In this work, we re-examined this problem using a new profile likelihood-based transformation estimation approach that is more statistically elegant and computationally efficient. We demonstrate the applicability of the proposed method using a benchmark Affymetrix U95A spiked-in experiment. Moreover, We introduced a new multivariate expression index and used the empirical study to shows its promise in terms of improving model fitting and power of detecting differential expression over the commonly used univariate expression index. As the other important content of the work, we discussed two generally encountered practical issues in application of gene expression index: normalization and summary statistic used for detecting differential expression. Our empirical study shows somewhat different findings from the MAQC project ( MAQC, 2006 ).
- 29 May 2012
- Published by Taylor & Francis in Journal of the American Statistical Association
- Vol. 107 (499) , 935-945
Searching for Alternative Splicing With a Joint Model on Probe Measurability and Expression Intensities
The exon tiling array offers a high throughput technology to search for aberrant splicing in biomedical research, but few methods of analysis for splicing detection have been tested both statistically and empirically. Noisy measurements on nonresponsive probe selection regions and outlying intensities at some of the samples tend to distort model-based assessments. We propose a robust analysis of variance approach that incorporates an informative model on probe measurability and uses median regression rank scores for better reliability in alternative splicing detection. We study the validity and effectiveness of our proposed approach in contrast with some of the existing methods through an empirical investigation of a brain cancer experiment, where a set of biologically validated genes for splicing and nonsplicing are available. Our study demonstrates favorable performance of the proposed ranking method, but shows that analysis of statistical significance cannot be trusted from any conventional use of p-values. We warn of any routine attempt to interpret p-values and their derivatives in model-based detection of alternative splicing.
- 29 May 2012
- Published by Taylor & Francis in Journal of the American Statistical Association
- Vol. 107 (499) , 935-945
Searching for Alternative Splicing With a Joint Model on Probe Measurability and Expression Intensities
The exon tiling array offers a high throughput technology to search for aberrant splicing in biomedical research, but few methods of analysis for splicing detection have been tested both statistically and empirically. Noisy measurements on nonresponsive probe selection regions and outlying intensities at some of the samples tend to distort model-based assessments. We propose a robust analysis of variance approach that incorporates an informative model on probe measurability and uses median regression rank scores for better reliability in alternative splicing detection. We study the validity and effectiveness of our proposed approach in contrast with some of the existing methods through an empirical investigation of a brain cancer experiment, where a set of biologically validated genes for splicing and nonsplicing are available. Our study demonstrates favorable performance of the proposed ranking method, but shows that analysis of statistical significance cannot be trusted from any conventional use of p-values. We warn of any routine attempt to interpret p-values and their derivatives in model-based detection of alternative splicing.
- Open Access
- 30 March 2012
- Published by Oxford University Press (OUP) in Bioinformatics
- Vol. 28 (9) , 1223-1229
Reno: regularized non-parametric analysis of protein lysate array data
Motivation: The reverse-phase protein lysate arrays have been used to quantify the relative expression levels of a protein in a number of cellular samples simultaneously. To avoid quantification bias due to mis-specification of commonly used parametric models, a nonparametric approach based on monotone response curves may be used. The existing methods, however, aggregate the protein concentration levels of replicates of each sample, and therefore fail to account for within-sample variability. Results: We propose a method of regularization on protein concentration estimation at the level of individual dilution series to account for within-sample or within-group variability. We use an efficient algorithm to optimize an approximate objective function, with a data-adaptive approach to choose the level of shrinkage. Simulation results show that the proposed method quantifies protein concentration levels well. We show through the analysis of protein lysate array data from cell lines of different cancer groups that accounting for within-sample variability leads to better statistical analysis. Availability: Code written in statistical programming language R is available at: http://odin.mdacc.tmc.edu/~jhhu/Reno Contact:jhu@mdanderson.org Supplementary information:Supplementary data are available at Bioinformatics online.
- 30 March 2012
- Published by Oxford University Press (OUP) in Bioinformatics
- Vol. 28 (9) , 1223-1229
Open AccessReno: regularized non-parametric analysis of protein lysate array data
Motivation: The reverse-phase protein lysate arrays have been used to quantify the relative expression levels of a protein in a number of cellular samples simultaneously. To avoid quantification bias due to mis-specification of commonly used parametric models, a nonparametric approach based on monotone response curves may be used. The existing methods, however, aggregate the protein concentration levels of replicates of each sample, and therefore fail to account for within-sample variability. Results: We propose a method of regularization on protein concentration estimation at the level of individual dilution series to account for within-sample or within-group variability. We use an efficient algorithm to optimize an approximate objective function, with a data-adaptive approach to choose the level of shrinkage. Simulation results show that the proposed method quantifies protein concentration levels well. We show through the analysis of protein lysate array data from cell lines of different cancer groups that accounting for within-sample variability leads to better statistical analysis. Availability: Code written in statistical programming language R is available at: http://odin.mdacc.tmc.edu/~jhhu/Reno Contact:jhu@mdanderson.org Supplementary information:Supplementary data are available at Bioinformatics online.