Jizhen Lin

How many replicates of arrays are required to detect gene expression changes in microarray experiments? A mixture model approach

BackgroundIt has been recognized that replicates of arrays (or spots) may be necessary for reliably detecting differentially expressed genes in microarray experiments. However, the often-asked question of how many replicates are required has barely been addressed in the literature. In general, the answer depends on several factors: a given magnitude of expression change, a desired statistical power (that is, probability) to detect it, a specified Type I error rate, and the statistical method being used to detect the change. Here, we discuss how to calculate the number of replicates in the context of applying a nonparametric statistical method, the normal mixture model approach, to detect changes in gene expression.ResultsThe methodology is applied to a data set containing expression levels of 1,176 genes in rats with and without pneumococcal middle-ear infection. We illustrate how to calculate the power functions for 2, 4, 6 and 8 replicates.ConclusionsThe proposed method is potentially useful in designing microarray experiments to discover differentially expressed genes. The same idea can be applied to other statistical methods.

A mixture model approach to detecting differentially expressed genes with microarray data

An exciting biological advancement over the past few years is the use of microarray technologies to measure simultaneously the expression levels of thousands of genes. The bottleneck now is how to extract useful information from the resulting large amounts of data. An important and common task in analyzing microarray data is to identify genes with altered expression under two experimental conditions. We propose a nonparametric statistical approach, called the mixture model method (MMM), to handle the problem when there are a small number of replicates under each experimental condition. Specifically, we propose estimating the distributions of a t -type test statistic and its null statistic using finite normal mixture models. A comparison of these two distributions by means of a likelihood ratio test, or simply using the tail distribution of the null statistic, can identify genes with significantly changed expression. Several methods are proposed to effectively control the false positives. The methodology is applied to a data set containing expression levels of 1,176 genes of rats with and without pneumococcal middle ear infection.

Model-based cluster analysis of microarray gene-expression data

BackgroundMicroarray technologies are emerging as a promising tool for genomic studies. The challenge now is how to analyze the resulting large amounts of data. Clustering techniques have been widely applied in analyzing microarray gene-expression data. However, normal mixture model-based cluster analysis has not been widely used for such data, although it has a solid probabilistic foundation. Here, we introduce and illustrate its use in detecting differentially expressed genes. In particular, we do not cluster gene-expression patterns but a summary statistic, the t-statistic.ResultsThe method is applied to a data set containing expression levels of 1,176 genes of rats with and without pneumococcal middle-ear infection. Three clusters were found, two of which contain more than 95% genes with almost no altered gene-expression levels, whereas the third one has 30 genes with more or less differential gene-expression levels.ConclusionsOur results indicate that model-based clustering of t-statistics (and possibly other summary statistics) can be a useful statistical tool to exploit differential gene expression for microarray data.

The role of inflammatory mediators in the pathogenesis of otitis media and sequelae.

This review deals with the characteristics of various inflammatory mediators identified in the middle ear during otitis media and in cholesteatoma. The role of each inflammatory mediator in the pathogenesis of otitis media and cholesteatoma has been discussed. Further, the relation of each inflammatory mediator to the pathophysiology of the middle and inner ear along with its mechanisms of pathological change has been described. The mechanisms of hearing loss including sensorineural hearing loss (SNHL) as a sequela of otitis media are also discussed. The passage of inflammatory mediators through the round window membrane into the scala tympani is indicated. In an experimental animal model, an application of cytokines and lipopolysaccharide (LPS), a bacterial toxin, on the round window membrane induced sensorineural hearing loss as identified through auditory brainstem response threshold shifts. An increase in permeability of the blood-labyrinth barrier (BLB) was observed following application of these inflammatory mediators and LPS. The leakage of the blood components into the lateral wall of the cochlea through an increase in BLB permeability appears to be related to the sensorineural hearing loss by hindering K+ recycling through the lateral wall disrupting the ion homeostasis of the endolymph. Further studies on the roles of various inflammatory mediators and bacterial toxins in inducing the sensorineumral hearing loss in otitis media should be pursued.

Inhibitor of differentiation 1 contributes to head and neck squamous cell carcinoma survival via the NF-κB/survivin and phosphoinositide 3-kinase/Akt signaling pathways

Purpose: A key issue in cancer is apoptosis resistance. However, little is known about the transcription factors that contribute to cellular survival of head and neck squamous cell carcinoma (HNSCC). Experimental Design: Three batches (54, 64, and 38) of HNSCC specimens were used for cellular and molecular analyses to determine the major molecular signaling pathways for cellular survival in HNSCC. Animal models (cell culture and xenografts) were used to verify the importance of apoptosis resistance in HNSCC. Results: Inhibitor of differentiation (Id) family member, Id1, was significantly upregulated in clinical HNSCC specimens and acted to protect keratinocytes from apoptosis. Transfection of HNSCC cells with Id1 in vitro induced the phosphorylation of Akt (p-Akt) via phosphoinositide 3-kinase and increased the expression of survivin via NF-κB. Blockage of both pathways by specific inhibitors (LY294002 and IκBαM, respectively) abrogated Id1-induced cell survival of keratinocytes. In vivo studies showed that increased expression of Id1 allowed nontumorigenic keratinocytes (Rhek-1A) to become tumorigenic in nude mice by increased expression of survival genes such as p-Akt and survivin. More importantly, short interfering RNA for Id1 significantly reduced HNSCC tumor volume of HNSCC in xenograft studies. Analysis of clinical data verified the importance of the Id1 downstream molecule, survivin, in the prognosis of HNSCC patients. Conclusions: The above data, taken together, suggest that Id1 and its downstream effectors are potential targets for treatment of HNSCC because of their contribution to apoptosis resistance. Clin Cancer Res; 16(1); 77–87

Effects of Inflammatory Mediators on Middle Ear Pathology and on Inner Ear Function

Inflammatory mediators released in the middle-ear cavity appear to play an important role in the pathogenesis of otitis media (OM) and can cause functional as well as morphological changes in the inner ear. Inflammatory mediators can be defined as biochemical components (peptide, glycoproteins, phospholipids, and others) produced by epithelial cells, infiltrating inflammatory cells, and endothelial cells, which mediate inflammatory reactions in a sequential manner. In the beginning, the only known target was microcirculation. In the late 1930s, Menkin’ showed that the accumulation of leukocytes could be explained in terms of mediators. As the number of known mediators has grown, the number of targets has also grown. At the present time, any cell is known to be a fair target of mediators, including epithelial cells, mast cells, fibroblasts, smooth muscles, endothelium, and white cells. As the number of known cell targets has increased, so has the number of possible cell responses, because every cell, when stimulated, will react according to its particular receptors and metabolism. All cells are known to exist in two forms, namely, resting and activated. Activated cells can produce many materials, including new mediators. According to Majno et a].: inflammatory mediators represent the basic language of cells. Cells must communicate through chemical messengers, whether they are resting or activated. The reason for so many mediators may be that

Coordinated Muc2 and Muc3 Mucin Gene Expression in Trichinella spiralis Infection in Wild-Type and Cytokine-Deficient Mice

Mucin hypersecretion is an important component of the immune response to gastrointestinal nematode infection. Two discrete types of mucin proteins exist in the mouse intestine, secretory Muc2 and membrane-bound Muc3. We examined Muc2 and Muc3 expression in wild-type mice and mice lacking gamma interferon receptor (IFNγR−/−), tumor necrosis factor receptor 1 (TNFR1−/−) and interleukin 4 (IL4−/−) infected with Trichinella spiralis. Infected wild-type mice demonstrated significant goblet cell hyperplasia and increased mucin glycoprotein. In situ hybridization showed this was accompanied by increases in Muc2 and Muc3 mRNA. Total intestinal mucin protein and Muc2 and Muc3 mRNA levels were also significantly increased in cytokine-deficient mice. These data demonstrate the coordinated up-regulation of two types of mucin genes in response to T. spiralis infection and may form the basis of an innate mucosal response independent of IFN-γ, TNF, and IL-4.

Identification of MUC5B mucin gene in human middle ear with chronic otitis media.

Objectives To identify the mucin gene and its expressing cells in the middle ear mucosa with chronic otitis media (COM), and to study the correlation between infiltration of inflammatory cells in the submucosa and expression of the mucin gene in the mucosal epithelium with COM.

Expression of Mucins in Mucoid Otitis Media

A hallmark of mucoid otitis media (MOM, i.e., chronic otitis media with mucoid effusion) is mucus accumulation in the middle ear cavity, a condition that impairs transduction of sounds in the ear and causes hearing loss. The mucin identities of mucus and the underlying mechanism for the production of mucins in MOM are poorly understood. In this study, we demonstrated that the MUC5B and MUC4 were major mucins in MOM that formed distinct treelike polymers (mucus strands). The MUC5B and MUC4 mRNAs in the middle ear mucosa with MOM were up regulated 5-fold and 6-fold, compared with the controls. This upregulation was accompanied by the extensive proliferation of the MUC5B- and MUC4-producing cells in the middle ear epithelium. Further study indicated that the mucin hyperproduction was significantly linked to CD4+ and CD8+ T cells and/or CD68+ monocyte macrophages. It suggests that MUC5B and MUC4 expression may be regulated by the products of these cells.

Cluster analysis using multivariate normal mixture models to detect differential gene expression with microarray data

DNA microarrays make it possible to study simultaneously the expression of thousands of genes in a biological sample. Univariate clustering techniques have been used to discover target genes with differential expression between two experimental conditions. Because of possible loss of information due to use of univariate summary statistics, it may be more effective to use multivariate statistics. We present multivariate normal mixture model based clustering analyses to detect differential gene expression between two conditions. Deviating from the general mixture model and model-based clustering, we propose mixture models with specific mean and covariance structures that account for special features of two-condition microarray experiments. Explicit updating formulas in the EM algorithm for three such models are derived. The methods are applied to a real dataset to compare the expression levels of 1176 genes of rats with and without pneumococcal middle-ear infection to illustrate the performance and usefulness of this approach. About 10 genes and 20 genes are found to be differentially expressed in a six-dimensional modeling and a bivariate modeling, respectively. Two simulation studies are conducted to compare the performance of univariate and multivariate methods. Depending on data, neither method can always dominate the other. The results suggest that multivariate normal mixture models can be useful alternatives to univariate methods to detect differential gene expression in exploratory data analysis.