Differential gene expression detection using penalized linear regression models: the improved SAM statistics

SUMMARY: Differential gene expression detection using microarrays has received lots of research interests recently. Many methods have been proposed, including variants of F-statistics, non-parametric approaches and empirical Bayesian methods etc. The SAM statistics has been shown to have good performance in empirical studies. SAM is more like an ad hoc shrinkage method. The idea is that for small sample microarray data, it is often useful to pool information across genes to improve efficiency. Under Bayesian framework Smyth formally derived the test statistics with shrinkage using the hierarchical models. In this paper we cast differential gene expression detection in the familiar framework of linear regression model. Commonly used test statistics correspond to using least squares to estimate the regression parameters. Based on the vast literature of research on linear models, we can naturally consider other alternatives. Here we explore the penalized linear regression. We propose the penalized t-/F-statistics for two-class microarray data based on [Formula: see text] penalty. We will show that the penalized test statistics intuitively makes sense and through applications we illustrate its good performance. AVAILABILITY: Supplementary information including program codes, more detailed analysis results and R functions for the proposed methods can be found at http://www.biostat.umn.edu/~baolin/research CONTACT: baolin@biostat.umn.edu SUPPLEMENTARY INFORMATION: http://www.biostat.umn.edu/~baolin/research.     

Dimension reduction-based penalized logistic regression for cancer classification using microarray data

The use of penalized logistic regression for cancer classification using microarray expression data is presented. Two dimension reduction methods are respectively combined with the penalized logistic regression so that both the classification accuracy and computational speed are enhanced. Two other machine-learning methods, support vector machines and least-squares regression, have been chosen for comparison. It is shown that our methods have achieved at least equal or better results. They also have the advantage that the output probability can be explicitly given and the regression coefficients are easier to interpret. Several other aspects, such as the selection of penalty parameters and components, pertinent to the application of our methods for cancer classification are also discussed.     

Latent group detection in functional partially linear regression models

In this paper, we propose a functional partially linear regression model with latent group structures to accommodate the heterogeneous relationship between a scalar response and functional covariates. The proposed model is motivated by a salinity tolerance study of barley families, whose main objective is to detect salinity tolerant barley plants. Our model is flexible, allowing for heterogeneous functional coefficients while being efficient by pooling information within a group for estimation. We develop an algorithm in the spirit of the K-means clustering to identify latent groups of the subjects under study. We establish the consistency of the proposed estimator, derive the convergence rate and the asymptotic distribution, and develop inference procedures. We show by simulation studies that the proposed method has higher accuracy for recovering latent groups and for estimating the functional coefficients than existing methods. The analysis of the barley data shows that the proposed method can help identify groups of barley families with different salinity tolerant abilities.     

Dynamic models of gene expression and classification

Powerful new methods, like expression profiles using cDNA arrays, have been used to monitor changes in gene expression levels as a result of a variety of metabolic, xenobiotic or pathogenic challenges. This potentially vast quantity of data enables, in principle, the dissection of the complex genetic networks that control the patterns and rhythms of gene expression in the cell. Here we present a general approach to developing dynamic models for analyzing time series of whole genome expression. In this approach, a self-consistent calculation is performed that involves both linear and non-linear response terms for interrelating gene expression levels. This calculation uses singular value decomposition (SVD) not as a statistical tool but as a means of inverting noisy and near-singular matrices. The linear transition matrix that is determined from this calculation can be used to calculate the underlying network reflected in the data. This suggests a direct method of classifying genes according to their place in the resulting network. In addition to providing a means to model such a large multivariate system this approach can be used to reduce the dimensionality of the problem in a rational and consistent way, and suppress the strong noise amplification effects often encountered with expression profile data. Non-linear and higher-order Markov behavior of the network are also determined in this self-consistent method. In data sets from yeast, we calculate the Markov matrix and the gene classes based on the linear-Markov network. These results compare favorably with previously used methods like cluster analysis. Our dynamic method appears to give a broad and general framework for data analysis and modeling of gene expression arrays. Electronic Publication     

Classification of gene microarrays by penalized logistic regression

Classification of patient samples is an important aspect of cancer diagnosis and treatment. The support vector machine (SVM) has been successfully applied to microarray cancer diagnosis problems. However, one weakness of the SVM is that given a tumor sample, it only predicts a cancer class label but does not provide any estimate of the underlying probability. We propose penalized logistic regression (PLR) as an alternative to the SVM for the microarray cancer diagnosis problem. We show that when using the same set of genes, PLR and the SVM perform similarly in cancer classification, but PLR has the advantage of additionally providing an estimate of the underlying probability. Often a primary goal in microarray cancer diagnosis is to identify the genes responsible for the classification, rather than class prediction. We consider two gene selection methods in this paper, univariate ranking (UR) and recursive feature elimination (RFE). Empirical results indicate that PLR combined with RFE tends to select fewer genes than other methods and also performs well in both cross-validation and test samples. A fast algorithm for solving PLR is also described.     

Developing optimal prediction models for cancer classification using gene expression data

Microarrays can provide genome-wide expression patterns for various cancers, especially for tumor sub-types that may exhibit substantially different patient prognosis. Using such gene expression data, several approaches have been proposed to classify tumor sub-types accurately. These classification methods are not robust, and often dependent on a particular training sample for modelling, which raises issues in utilizing these methods to administer proper treatment for a future patient. We propose to construct an optimal, robust prediction model for classifying cancer sub-types using gene expression data. Our model is constructed in a step-wise fashion implementing cross-validated quadratic discriminant analysis. At each step, all identified models are validated by an independent sample of patients to develop a robust model for future data. We apply the proposed methods to two microarray data sets of cancer: the acute leukemia data by Golub et al. and the colon cancer data by Alon et al. We have found that the dimensionality of our optimal prediction models is relatively small for these cases and that our prediction models with one or two gene factors outperforms or has competing performance, especially for independent samples, to other methods based on 50 or more predictive gene factors. The methodology is implemented and developed by the procedures in R and Splus. The source code can be obtained at http://hesweb1.med.virginia.edu/bioinformatics.     

Linear regression and two-class classification with gene expression data

MOTIVATION: Using gene expression data to classify (or predict) tumor types has received much research attention recently. Due to some special features of gene expression data, several new methods have been proposed, including the weighted voting scheme of Golub et al., the compound covariate method of Hedenfalk et al. (originally proposed by Tukey), and the shrunken centroids method of Tibshirani et al. These methods look different and are more or less ad hoc. RESULTS: We point out a close connection of the three methods with a linear regression model. Casting the classification problem in the general framework of linear regression naturally leads to new alternatives, such as partial least squares (PLS) methods and penalized PLS (PPLS) methods. Using two real data sets, we show the competitive performance of our new methods when compared with the other three methods.     

Eigengene-based linear discriminant model for tumor classification using gene expression microarray data

MOTIVATION: The nearest shrunken centroids classifier has become a popular algorithm in tumor classification problems using gene expression microarray data. Feature selection is an embedded part of the method to select top-ranking genes based on a univariate distance statistic calculated for each gene individually. The univariate statistics summarize gene expression profiles outside of the gene co-regulation network context, leading to redundant information being included in the selection procedure. RESULTS: We propose an Eigengene-based Linear Discriminant Analysis (ELDA) to address gene selection in a multivariate framework. The algorithm uses a modified rotated Spectral Decomposition (SpD) technique to select 'hub' genes that associate with the most important eigenvectors. Using three benchmark cancer microarray datasets, we show that ELDA selects the most characteristic genes, leading to substantially smaller classifiers than the univariate feature selection based analogues. The resulting de-correlated expression profiles make the gene-wise independence assumption more realistic and applicable for the shrunken centroids classifier and other diagonal linear discriminant type of models. Our algorithm further incorporates a misclassification cost matrix, allowing differential penalization of one type of error over another. In the breast cancer data, we show false negative prognosis can be controlled via a cost-adjusted discriminant function. AVAILABILITY: R code for the ELDA algorithm is available from author upon request.     

Classification using partial least squares with penalized logistic regression

MOTIVATION: One important aspect of data-mining of microarray data is to discover the molecular variation among cancers. In microarray studies, the number n of samples is relatively small compared to the number p of genes per sample (usually in thousands). It is known that standard statistical methods in classification are efficient (i.e. in the present case, yield successful classifiers) particularly when n is (far) larger than p. This naturally calls for the use of a dimension reduction procedure together with the classification one. RESULTS: In this paper, the question of classification in such a high-dimensional setting is addressed. We view the classification problem as a regression one with few observations and many predictor variables. We propose a new method combining partial least squares (PLS) and Ridge penalized logistic regression. We review the existing methods based on PLS and/or penalized likelihood techniques, outline their interest in some cases and theoretically explain their sometimes poor behavior. Our procedure is compared with these other classifiers. The predictive performance of the resulting classification rule is illustrated on three data sets: Leukemia, Colon and Prostate.     

Biologically valid linear factor models of gene expression

MOTIVATION: The identification of physiological processes underlying and generating the expression pattern observed in microarray experiments is a major challenge. Principal component analysis (PCA) is a linear multivariate statistical method that is regularly employed for that purpose as it provides a reduced-dimensional representation for subsequent study of possible biological processes responding to the particular experimental conditions. Making explicit the data assumptions underlying PCA highlights their lack of biological validity thus making biological interpretation of the principal components problematic. A microarray data representation which enables clear biological interpretation is a desirable analysis tool. RESULTS: We address this issue by employing the probabilistic interpretation of PCA and proposing alternative linear factor models which are based on refined biological assumptions. A practical study on two well-understood microarray datasets highlights the weakness of PCA and the greater biological interpretability of the linear models we have developed.     

Semi-supervised learning via penalized mixture model with application to microarray sample classification

MOTIVATION: It is biologically interesting to address whether human blood outgrowth endothelial cells (BOECs) belong to or are closer to large vessel endothelial cells (LVECs) or microvascular endothelial cells (MVECs) based on global expression profiling. An earlier analysis using a hierarchical clustering and a small set of genes suggested that BOECs seemed to be closer to MVECs. By taking advantage of the two known classes, LVEC and MVEC, while allowing BOEC samples to belong to either of the two classes or to form their own new class, we take a semi-supervised learning approach; for high-dimensional data as encountered here, we propose a penalized mixture model with a weighted L1 penalty to realize automatic feature selection while fitting the model. RESULTS: We applied our penalized mixture model to a combined dataset containing 27 BOEC, 28 LVEC and 25 MVEC samples. Analysis results indicated that the BOEC samples appeared to form their own new class. A simulation study confirmed that, compared with the standard mixture model with or without initial variable selection, the penalized mixture model performed much better in identifying relevant genes and forming corresponding clusters. The penalized mixture model seems to be promising for high-dimensional data with the capability of novel class discovery and automatic feature selection.     

基于核局部线性嵌入的基因表达谱数据分类

针对局部线性嵌入算法（LocalLinearEmbedding,LLE）利用试凑法寻找近邻数耗时的缺陷性,提出一种增强的核局部线性嵌入算法（EnhancedKernelLocalLinearEmbedding,EKLLE）自动为样本分配邻域;该算法以高斯核函数为核心改进标准LLE距离度量准则,结合样本的类别信息,无需人工干预自动为样本设置不同的近邻数,克服了试凑法获得最优结果时需要大量时间;最后在各样本近邻数不相同的情况下对数据进行维数简约及待测样本分类。EKLLE算法有效地将高维基因表达谱数据映射到低维本质空间中,解决了传统LLE算法不能很好地处理合噪声或者稀疏数据的缺点。通过对比其他肿瘤样本分类实验,验证本文方法的实时性和精确性。     

Random regression models for detection of gene by environment interaction

Two random regression models, where the effect of a putative QTL was regressed on an environmental gradient, are described. The first model estimates the correlation between intercept and slope of the random regression, while the other model restricts this correlation to 1 or -1, which is expected under a bi-allelic QTL model. The random regression models were compared to a model assuming no gene by environment interactions. The comparison was done with regards to the models ability to detect QTL, to position them accurately and to detect possible QTL by environment interactions. A simulation study based on a granddaughter design was conducted, and QTL were assumed, either by assigning an effect independent of the environment or as a linear function of a simulated environmental gradient. It was concluded that the random regression models were suitable for detection of QTL effects, in the presence and absence of interactions with environmental gradients. Fixing the correlation between intercept and slope of the random regression had a positive effect on power when the QTL effects re-ranked between environments.     

Missing-value estimation using linear and non-linear regression with Bayesian gene selection

MOTIVATION: Data from microarray experiments are usually in the form of large matrices of expression levels of genes under different experimental conditions. Owing to various reasons, there are frequently missing values. Estimating these missing values is important because they affect downstream analysis, such as clustering, classification and network design. Several methods of missing-value estimation are in use. The problem has two parts: (1) selection of genes for estimation and (2) design of an estimation rule. RESULTS: We propose Bayesian variable selection to obtain genes to be used for estimation, and employ both linear and nonlinear regression for the estimation rule itself. Fast implementation issues for these methods are discussed, including the use of QR decomposition for parameter estimation. The proposed methods are tested on data sets arising from hereditary breast cancer and small round blue-cell tumors. The results compare very favorably with currently used methods based on the normalized root-mean-square error. AVAILABILITY: The appendix is available from http://gspsnap.tamu.edu/gspweb/zxb/missing_zxb/ (user: gspweb; passwd: gsplab).     

Multi-class cancer classification by total principal component regression (TPCR) using microarray gene expression data

DNA microarray technology provides a promising approach to the diagnosis and prognosis of tumors on a genome-wide scale by monitoring the expression levels of thousands of genes simultaneously. One problem arising from the use of microarray data is the difficulty to analyze the high-dimensional gene expression data, typically with thousands of variables (genes) and much fewer observations (samples), in which severe collinearity is often observed. This makes it difficult to apply directly the classical statistical methods to investigate microarray data. In this paper, total principal component regression (TPCR) was proposed to classify human tumors by extracting the latent variable structure underlying microarray data from the augmented subspace of both independent variables and dependent variables. One of the salient features of our method is that it takes into account not only the latent variable structure but also the errors in the microarray gene expression profiles (independent variables). The prediction performance of TPCR was evaluated by both leave-one-out and leave-half-out cross-validation using four well-known microarray datasets. The stabilities and reliabilities of the classification models were further assessed by re-randomization and permutation studies. A fast kernel algorithm was applied to decrease the computation time dramatically. (MATLAB source code is available upon request.).