Zhenyu Jia

Mapping Quantitative Trait Loci for Expression Abundance

Mendelian loci that control the expression levels of transcripts are called expression quantitative trait loci (eQTL). When mapping eQTL, we often deal with thousands of expression traits simultaneously, which complicates the statistical model and data analysis. Two simple approaches may be taken in eQTL analysis: (1) individual transcript analysis in which a single expression trait is mapped at a time and the entire eQTL mapping involves separate analysis of thousands of traits and (2) individual marker analysis where differentially expressed transcripts are detected on the basis of their association with the segregation pattern of an individual marker and the entire analysis requires scanning markers of the entire genome. Neither approach is optimal because data are not analyzed jointly. We develop a Bayesian clustering method that analyzes all expressed transcripts and markers jointly in a single model. A transcript may be simultaneously associated with multiple markers. Additionally, a marker may simultaneously alter the expression of multiple transcripts. This is a model-based method that combines a Gaussian mixture of expression data with segregation of multiple linked marker loci. Parameter estimation for each variable is obtained via the posterior mean drawn from a Markov chain Monte Carlo sample. The method allows a regular quantitative trait to be included as an expression trait and subject to the same clustering assignment. If an expression trait links to a locus where a quantitative trait also links, the expressed transcript is considered to be associated with the quantitative trait. The method is applied to a microarray experiment with 60 F2 mice measured for 25 different obesity-related quantitative traits. In the experiment, ∼40,000 transcripts and 145 codominant markers are investigated for their associations. A program written in SAS/IML is available from the authors on request.

Differential Expression of Peroxiredoxins in Prostate Cancer: Consistent Upregulation of PRDX3 and PRDX4

The peroxiredoxins (PRDXs) are emerging as regulators of antioxidant defense, apoptosis, and therapy resistance in cancer. Because their significance in prostate cancer (PCa) is unclear, we investigated their expression and clinical associations in PCa.

Diagnosis of prostate cancer using differentially expressed genes in stroma.

More than one million prostate biopsies are performed in the United States every year. A failure to find cancer is not definitive in a significant percentage of patients due to the presence of equivocal structures or continuing clinical suspicion. We have identified gene expression changes in stroma that can detect tumor nearby. We compared gene expression profiles of 13 biopsies containing stroma near tumor and 15 biopsies from volunteers without prostate cancer. About 3,800 significant expression changes were found and thereafter filtered using independent expression profiles to eliminate possible age-related genes and genes expressed at detectable levels in tumor cells. A stroma-specific classifier for nearby tumor was constructed on the basis of 114 candidate genes and tested on 364 independent samples including 243 tumor-bearing samples and 121 nontumor samples (normal biopsies, normal autopsies, remote stroma, as well as stroma within a few millimeters of tumor). The classifier predicted the tumor status of patients using tumor-free samples with an average accuracy of 97% (sensitivity = 98% and specificity = 88%) whereas classifiers trained with sets of 100 randomly generated genes had no diagnostic value. These results indicate that the prostate cancer microenvironment exhibits reproducible changes useful for categorizing the presence of tumor in patients when a prostate sample is derived from near the tumor but does not contain any recognizable tumor.

Clustering expressed genes on the basis of their association with a quantitative phenotype

Cluster analyses of gene expression data are usually conducted based on their associations with the phenotype of a particular disease. Many disease traits have a clearly defined binary phenotype (presence or absence), so that genes can be clustered based on the differences of expression levels between the two contrasting phenotypic groups. For example, cluster analysis based on binary phenotype has been successfully used in tumour research. Some complex diseases have phenotypes that vary in a continuous manner and the method developed for a binary trait is not immediately applicable to a continuous trait. However, understanding the role of gene expression in these complex traits is of fundamental importance. Therefore, it is necessary to develop a new statistical method to cluster expressed genes based on their association with a quantitative trait phenotype. We developed a model-based clustering method to classify genes based on their association with a continuous phenotype. We used a linear model to describe the relationship between gene expression and the phenotypic value. The model effects of the linear model (linear regression coefficients) represent the strength of the association. We assumed that the model effects of each gene follow a mixture of several multivariate Gaussian distributions. Parameter estimation and cluster assignment were accomplished via an Expectation-Maximization (EM) algorithm. The method was verified by analysing two simulated datasets, and further demonstrated using real data generated in a microarray experiment for the study of gene expression associated with Alzheimers disease.

The Concordance Index C and the Mann―Whitney Parameter Pr(X>Y) with Randomly Censored Data

Harrells c-index or concordance C has been widely used as a measure of separation of two survival distributions. In the absence of censored data, the c-index estimates the Mann-Whitney parameter Pr(X>Y), which has been repeatedly utilized in various statistical contexts. In the presence of randomly censored data, the c-index no longer estimates Pr(X>Y); rather, a parameter that involves the underlying censoring distributions. This is in contrast to Efrons maximum likelihood estimator of the Mann-Whitney parameter, which is recommended in the setting of random censorship.

The wisdom of the commons

MOTIVATION Classification and regression trees have long been used for cancer diagnosis and prognosis. Nevertheless, instability and variable selection bias, as well as overfitting, are well-known problems of tree-based methods. In this article, we investigate whether ensemble tree classifiers can ameliorate these difficulties, using data from two recent studies of radical prostatectomy in prostate cancer. RESULTS Using time to progression following prostatectomy as the relevant clinical endpoint, we found that ensemble tree classifiers robustly and reproducibly identified three subgroups of patients in the two clinical datasets: non-progressors, early progressors and late progressors. Moreover, the consensus classifications were independent predictors of time to progression compared to known clinical prognostic factors.

Expression differences between African American and Caucasian prostate cancer tissue reveals that stroma is the site of aggressive changes

In prostate cancer, race/ethnicity is the highest risk factor after adjusting for age. African Americans have more aggressive tumors at every clinical stage of the disease, resulting in poorer prognosis and increased mortality. A major barrier to identifying crucial gene activity differences is heterogeneity, including tissue composition variation intrinsic to the histology of prostate cancer. We hypothesized that differences in gene expression in specific tissue types would reveal mechanisms involved in the racial disparities of prostate cancer. We examined 17 pairs of arrays for AAs and Caucasians that were formed by closely matching the samples based on the known tissue type composition of the tumors. Using pair‐wise t‐test we found significantly altered gene expression between AAs and CAs. Independently, we performed multiple linear regression analyses to associate gene expression with race considering variation in percent tumor and stroma tissue. The majority of differentially expressed genes were associated with tumor‐adjacent stroma rather than tumor tissue. Extracellular matrix, integrin family and signaling mediators of the epithelial‐to‐mesenchymal transition (EMT) pathways were all downregulated in stroma of AAs. Using MetaCore (GeneGo) analysis, we observed that 35% of significant (p < 10−3) pathways identified EMT and 25% identified immune response pathways especially for interleukins‐2, ‐4, ‐5, ‐6, ‐7, ‐10, ‐13, ‐15 and ‐22 as the major changes. Our studies reveal that altered immune and EMT processes in tumor‐adjacent stroma may be responsible for the aggressive nature of prostate cancer in AAs.

Egr1 regulates the coordinated expression of numerous EGF receptor target genes as identified by ChIP-on-chip.

BackgroundUV irradiation activates the epidermal growth factor receptor, induces Egr1 expression and promotes apoptosis in a variety of cell types. We examined the hypothesis that Egr1 regulates genes that mediate this process by use of a chip-on-chip protocol in human tumorigenic prostate M12 cells.ResultsUV irradiation led to significant binding of 288 gene promoters by Egr1. A major functional subgroup consisted of apoptosis related genes. The largest subgroup of 24 genes belongs to the epidermal growth factor receptor-signal transduction pathway. Egr1 promoter binding had a significant impact on gene expression of target genes. Conventional chromatin immunoprecipitation and quantitative real time PCR were used to validate promoter binding and expression changes. Small interfering RNA experiments were used to demonstrate the specific role of Egr1 in gene regulation. UV stimulation promotes growth arrest and apoptosis of M12 cells and our data clearly show that a downstream target of the epidermal growth factor receptor, namely Egr1, mediates this apoptotic response. Our study also identified numerous previously unknown targets of Egr1. These include FasL, MAX and RRAS2, which may play a role in the apoptotic response/growth arrest.ConclusionsOur results indicate that M12 cells undergo Egr1-dependent apoptotic response upon UV stimulation and led to the identification of downstream targets of Egr1, which mediate epidermal growth factor receptor function.

Expression Changes in the Stroma of Prostate Cancer Predict Subsequent Relapse

Biomarkers are needed to address overtreatment that occurs for the majority of prostate cancer patients that would not die of the disease but receive radical treatment. A possible barrier to biomarker discovery may be the polyclonal/multifocal nature of prostate tumors as well as cell-type heterogeneity between patient samples. Tumor-adjacent stroma (tumor microenvironment) is less affected by genetic alteration and might therefore yield more consistent biomarkers in response to tumor aggressiveness. To this end we compared Affymetrix gene expression profiles in stroma near tumor and identified a set of 115 probe sets for which the expression levels were significantly correlated with time-to-relapse. We also compared patients that chemically relapsed shortly after prostatectomy (<1 year), and patients that did not relapse in the first four years after prostatectomy. We identified 131 differentially expressed microarray probe sets between these two categories. 19 probe sets (15 genes overlapped between the two gene lists with p<0.0001). We developed a PAM-based classifier by training on samples containing stroma near tumor: 9 rapid relapse patient samples and 9 indolent patient samples. We then tested the classifier on 47 different samples, containing 90% or more stroma. The classifier predicted the risk status of patients with an average accuracy of 87%. This is the first general tumor microenvironment-based prognostic classifier. These results indicate that the prostate cancer microenvironment exhibits reproducible changes useful for predicting outcomes for patients.

An Accurate Prostate Cancer Prognosticator Using a Seven-Gene Signature Plus Gleason Score and Taking Cell Type Heterogeneity into Account

One of the major challenges in the development of prostate cancer prognostic biomarkers is the cellular heterogeneity in tissue samples. We developed an objective Cluster-Correlation (CC) analysis to identify gene expression changes in various cell types that are associated with progression. In the Cluster step, samples were clustered (unsupervised) based on the expression values of each gene through a mixture model combined with a multiple linear regression model in which cell-type percent data were used for decomposition. In the Correlation step, a Chi-square test was used to select potential prognostic genes. With CC analysis, we identified 324 significantly expressed genes (68 tumor and 256 stroma cell expressed genes) which were strongly associated with the observed biochemical relapse status. Significance Analysis of Microarray (SAM) was then utilized to develop a seven-gene classifier. The Classifier has been validated using two independent Data Sets. The overall prediction accuracy and sensitivity is 71% and 76%, respectively. The inclusion of the Gleason sum to the seven-gene classifier raised the prediction accuracy and sensitivity to 83% and 76% respectively based on independent testing. These results indicated that our prognostic model that includes cell type adjustments and using Gleason score and the seven-gene signature has some utility for predicting outcomes for prostate cancer for individual patients at the time of prognosis. The strategy could have applications for improving marker performance in other cancers and other diseases.