Multi-class cancer subtype classification based on gene expression signatures with reliability analysis

Differential diagnosis among a group of histologically similar cancers poses a challenging problem in clinical medicine. Constructing a classifier based on gene expression signatures comprising multiple discriminatory molecular markers derived from microarray data analysis is an emerging trend for cancer diagnosis. To identify the best genes for classification using a small number of samples relative to the genome size remains the bottleneck of this approach, despite its promise. We have devised a new method of gene selection with reliability analysis, and demonstrated that this method can identify a more compact set of genes than other methods for constructing a classifier with optimum predictive performance for both small round blue cell tumors and leukemia. High consensus between our result and the results produced by methods based on artificial neural networks and statistical techniques confers additional evidence of the validity of our method. This study suggests a way for implementing a reliable molecular cancer classifier based on gene expression signatures.     

Peripheral blood gene expression profiles in COPD subjects

To identify non-invasive gene expression markers for chronic obstructive pulmonary disease (COPD), we performed genome-wide expression profiling of peripheral blood samples from 12 subjects with significant airflow obstruction and an equal number of non-obstructed controls. RNA was isolated from Peripheral Blood Mononuclear Cells (PBMCs) and gene expression was assessed using Affymetrix U133 Plus 2.0 arrays.Tests for gene expression changes that discriminate between COPD cases (FEV1< 70% predicted, FEV1/FVC < 0.7) and controls (FEV1> 80% predicted, FEV1/FVC > 0.7) were performed using Significance Analysis of Microarrays (SAM) and Bayesian Analysis of Differential Gene Expression (BADGE). Using either test at high stringency (SAM median FDR = 0 or BADGE p < 0.01) we identified differential expression for 45 known genes. Correlation of gene expression with lung function measurements (FEV1 & FEV1/FVC), using both Pearson and Spearman correlation coefficients (p < 0.05), identified a set of 86 genes. A total of 16 markers showed evidence of significant correlation (p < 0.05) with quantitative traits and differential expression between cases and controls. We further compared our peripheral gene expression markers with those we previously identified from lung tissue of the same cohort. Two genes, RP9and NAPE-PLD, were identified as decreased in COPD cases compared to controls in both lung tissue and blood. These results contribute to our understanding of gene expression changes in the peripheral blood of patients with COPD and may provide insight into potential mechanisms involved in the disease.     

Peripheral blood gene expression profiling for cardiovascular disease assessment

Whole blood gene expression profiling has the potential to be informative about dynamic changes in disease states and to provide information on underlying disease mechanisms. Having demonstrated proof of concept in animal models, a number of studies have now tried to tackle the complexity of cardiovascular disease in human hosts to develop better diagnostic and prognostic indicators. These studies show that genomic signatures are capable of classifying patients with cardiovascular diseases into finer categories based on the molecular architecture of a patient's disease and more accurately predict the likelihood of a cardiovascular event than current techniques. To highlight the spectrum of potential applications of whole blood gene expression profiling approach in cardiovascular science, we have chosen to review the findings in a number of complex cardiovascular diseases such as atherosclerosis, hypertension and myocardial infarction as well as thromboembolism, aortic aneurysm, and heart transplant.     

Shared signatures of social stress and aging in peripheral blood mononuclear cell gene expression profiles

Chronic social stress is a predictor of both aging‐related disease and mortality risk. Hence, chronic stress has been hypothesized to directly exacerbate the process of physiological aging. Here, we evaluated this hypothesis at the level of gene regulation. We compared two data sets of genome‐wide gene expression levels in peripheral blood mononuclear cells (PBMCs): one that captured aging effects and another that focused on chronic social stress. Overall, we found that the direction, although not necessarily the magnitude, of significant gene expression changes tends to be shared between the two data sets. This overlap was observable at three levels: (i) individual genes; (ii) general functional categories of genes; and (iii) molecular pathways implicated in aging. However, we also found evidence that heterogeneity in PBMC composition limits the power to detect more extensive similarities, suggesting that our findings reflect an underestimate of the degree to which age and social stress influence gene regulation in parallel. Cell type‐specific data on gene regulation will be important to overcome this limitation in the future studies.     

Peripheral blood leukocytes of cows with subclinical endometritis show an altered cellular composition and gene expression

Subclinical endometritis (SCE) is an important postpartum disease in dairy cows, but conventional cytobrush diagnosis often gives imprecise results. The aim of this study was to analyze disease-associated changes in peripheral blood as potential diagnostic parameters. Cellular subpopulations of blood leukocytes from cows with or without SCE (45–55 days postpartum) were flow-cytometrically quantified. Gene expression of whole blood leukocytes was assessed by PAXgene analysis. Subclinical endometritis cows showed significantly higher number of blood mononuclear cells and neutrophils. Among mononuclear cells, numbers of B-cells, NK-cells, and CD172a-positive monocytes were significantly elevated. Compared with non-SCE cows, blood leukocytes of SCE cows significantly expressed higher copy numbers of CXCL8, TNF, and IL12. To test whether circulating plasma factors are responsible for these changes, leukocytes, polymorphonuclear cells, and monocyte subpopulations (classical, intermediate, nonclassical) of healthy cows were stimulated with plasma of SCE and non-SCE cows. Although gene expression of whole leukocytes and polymorphonuclear cells remained unaltered, plasma from SCE animals significantly elevated expressed messenger RNA copy numbers of CXCL8, CXCL1, and IL1B in intermediate monocytes. In conclusion, elevated number of selected mononuclear subpopulations in peripheral blood and enhanced expression of distinct genes encoding for inflammatory mediators in blood leukocytes reflect the subclinical uterine inflammatory process in cows. Whether the observed changes in the periphery of SCE cows are the consequence of the uterine inflammatory process, or whether they affect the pathogenesis of the disease is currently unknown.     

A decomposition model to track gene expression signatures: preview on observer-independent classification of ovarian cancer

MOTIVATION: A number of algorithms and analytical models have been employed to reduce the multidimensional complexity of DNA array data and attempt to extract some meaningful interpretation of the results. These include clustering, principal components analysis, self-organizing maps, and support vector machine analysis. Each method assumes an implicit model for the data, many of which separate genes into distinct clusters defined by similar expression profiles in the samples tested. A point of concern is that many genes may be involved in a number of distinct behaviours, and should therefore be modelled to fit into as many separate clusters as detected in the multidimensional gene expression space. The analysis of gene expression data using a decomposition model that is independent of the observer involved would be highly beneficial to improve standard and reproducible classification of clinical and research samples. RESULTS: We present a variational independent component analysis (ICA) method for reducing high dimensional DNA array data to a smaller set of latent variables, each associated with a gene signature. We present the results of applying the method to data from an ovarian cancer study, revealing a number of tissue type-specific and tissue type-independent gene signatures present in varying amounts among the samples surveyed. The observer independent results of such molecular analysis of biological samples could help identify patients who would benefit from different treatment strategies. We further explore the application of the model to similar high-throughput studies.     

Tissue classification with gene expression profiles.

Constantly improving gene expression profiling technologies are expected to provide understanding and insight into cancer-related cellular processes. Gene expression data is also expected to significantly aid in the development of efficient cancer diagnosis and classification platforms. In this work we examine three sets of gene expression data measured across sets of tumor(s) and normal clinical samples: The first set consists of 2,000 genes, measured in 62 epithelial colon samples (Alon et al., 1999). The second consists of approximately equal to 100,000 clones, measured in 32 ovarian samples (unpublished extension of data set described in Schummer et al. (1999)). The third set consists of approximately equal to 7,100 genes, measured in 72 bone marrow and peripheral blood samples (Golub et al, 1999). We examine the use of scoring methods, measuring separation of tissue type (e.g., tumors from normals) using individual gene expression levels. These are then coupled with high-dimensional classification methods to assess the classification power of complete expression profiles. We present results of performing leave-one-out cross validation (LOOCV) experiments on the three data sets, employing nearest neighbor classifier, SVM (Cortes and Vapnik, 1995), AdaBoost (Freund and Schapire, 1997) and a novel clustering-based classification technique. As tumor samples can differ from normal samples in their cell-type composition, we also perform LOOCV experiments using appropriately modified sets of genes, attempting to eliminate the resulting bias. We demonstrate success rate of at least 90% in tumor versus normal classification, using sets of selected genes, with, as well as without, cellular-contamination-related members. These results are insensitive to the exact selection mechanism, over a certain range.     

Efficient human iPS cell derivation by a non-integrating plasmid from blood cells with unique epigenetic and gene expression signatures

To identify accessible and permissive human cell types for efficient derivation of induced pluripotent stem cells (iPSCs), we investigated epigenetic and gene expression signatures of multiple postnatal cell types such as fibroblasts and blood cells. Our analysis suggested that newborn cord blood (CB) and adult peripheral blood (PB) mononuclear cells (MNCs) display unique signatures that are closer to iPSCs and human embryonic stem cells (ESCs) than age-matched fibroblasts to iPSCs/ESCs, thus making blood MNCs an attractive cell choice for the generation of integration-free iPSCs. Using an improved EBNA1/OriP plasmid expressing 5 reprogramming factors, we demonstrated highly efficient reprogramming of briefly cultured blood MNCs. Within 14 days of one-time transfection by one plasmid, up to 1000 iPSC-like colonies per 2 million transfected CB MNCs were generated. The efficiency of deriving iPSCs from adult PB MNCs was approximately 50-fold lower, but could be enhanced by inclusion of a second EBNA1/OriP plasmid for transient expression of additional genes such as SV40 T antigen. The duration of obtaining bona fide iPSC colonies from adult PB MNCs was reduced to half (～14 days) as compared to adult fibroblastic cells (28-30 days). More than 9 human iPSC lines derived from PB or CB blood cells are extensively characterized, including those from PB MNCs of an adult patient with sickle cell disease. They lack V(D)J DNA rearrangements and vector DNA after expansion for 10-12 passages. This facile method of generating integration-free human iPSCs from blood MNCs will accelerate their use in both research and future clinical applications.     

Strategy for encoding and comparison of gene expression signatures

EXALT (EXpression signature AnaLysis Tool) is a computational system enabling comparisons of microarray data across experimental platforms and different laboratories . An essential feature of EXALT is a database holding thousands of gene expression signatures extracted from the Gene Expression Omnibus, and encoded in a searchable format. This novel approach to performing global comparisons of shared microarray data may have enormous value when coupled directly with a shared data repository.     

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.     

Mining gene expression profiles: expression signatures as cancer phenotypes

Many examples highlight the power of gene expression profiles, or signatures, to inform an understanding of biological phenotypes. This is perhaps best seen in the context of cancer, where expression signatures have tremendous power to identify new subtypes and to predict clinical outcomes. Although the ability to interpret the meaning of the individual genes in these signatures remains a challenge, this does not diminish the power of the signature to characterize biological states. The use of these signatures as surrogate phenotypes has been particularly important, linking diverse experimental systems that dissect the complexity of biological systems with the in vivo setting in a way that was not previously feasible.     

Reliable gene signatures for microarray classification: assessment of stability and performance

MOTIVATION: Two important questions for the analysis of gene expression measurements from different sample classes are (1) how to classify samples and (2) how to identify meaningful gene signatures (ranked gene lists) exhibiting the differences between classes and sample subsets. Solutions to both questions have immediate biological and biomedical applications. To achieve optimal classification performance, a suitable combination of classifier and gene selection method needs to be specifically selected for a given dataset. The selected gene signatures can be unstable and the resulting classification accuracy unreliable, particularly when considering different subsets of samples. Both unstable gene signatures and overestimated classification accuracy can impair biological conclusions. METHODS: We address these two issues by repeatedly evaluating the classification performance of all models, i.e. pairwise combinations of various gene selection and classification methods, for random subsets of arrays (sampling). A model score is used to select the most appropriate model for the given dataset. Consensus gene signatures are constructed by extracting those genes frequently selected over many samplings. Sampling additionally permits measurement of the stability of the classification performance for each model, which serves as a measure of model reliability. RESULTS: We analyzed a large gene expression dataset with 78 measurements of four different cartilage sample classes. Classifiers trained on subsets of measurements frequently produce models with highly variable performance. Our approach provides reliable classification performance estimates via sampling. In addition to reliable classification performance, we determined stable consensus signatures (i.e. gene lists) for sample classes. Manual literature screening showed that these genes are highly relevant to our gene expression experiment with osteoarthritic cartilage. We compared our approach to others based on a publicly available dataset on breast cancer. AVAILABILITY: R package at http://www.bio.ifi.lmu.de/~davis/edaprakt     

Web-based interrogation of gene expression signatures using EXALT

Background  

Widespread use of high-throughput techniques such as microarrays to monitor gene expression levels has resulted in an explosive growth of data sets in public domains. Integration and exploration of these complex and heterogeneous data have become a major challenge.     

BagBoosting for tumor classification with gene expression data

MOTIVATION: Microarray experiments are expected to contribute significantly to the progress in cancer treatment by enabling a precise and early diagnosis. They create a need for class prediction tools, which can deal with a large number of highly correlated input variables, perform feature selection and provide class probability estimates that serve as a quantification of the predictive uncertainty. A very promising solution is to combine the two ensemble schemes bagging and boosting to a novel algorithm called BagBoosting. RESULTS: When bagging is used as a module in boosting, the resulting classifier consistently improves the predictive performance and the probability estimates of both bagging and boosting on real and simulated gene expression data. This quasi-guaranteed improvement can be obtained by simply making a bigger computing effort. The advantageous predictive potential is also confirmed by comparing BagBoosting to several established class prediction tools for microarray data. AVAILABILITY: Software for the modified boosting algorithms, for benchmark studies and for the simulation of microarray data are available as an R package under GNU public license at http://stat.ethz.ch/~dettling/bagboost.html.     

Boosting for tumor classification with gene expression data

MOTIVATION: Microarray experiments generate large datasets with expression values for thousands of genes but not more than a few dozens of samples. Accurate supervised classification of tissue samples in such high-dimensional problems is difficult but often crucial for successful diagnosis and treatment. A promising way to meet this challenge is by using boosting in conjunction with decision trees. RESULTS: We demonstrate that the generic boosting algorithm needs some modification to become an accurate classifier in the context of gene expression data. In particular, we present a feature preselection method, a more robust boosting procedure and a new approach for multi-categorical problems. This allows for slight to drastic increase in performance and yields competitive results on several publicly available datasets. AVAILABILITY: Software for the modified boosting algorithms as well as for decision trees is available for free in R at http://stat.ethz.ch/~dettling/boosting.html.     

Quantitative comparison of microarray experiments with published leukemia related gene expression signatures

Background  

Multiple gene expression signatures derived from microarray experiments have been published in the field of leukemia research. A comparison of these signatures with results from new experiments is useful for verification as well as for interpretation of the results obtained. Currently, the percentage of overlapping genes is frequently used to compare published gene signatures against a signature derived from a new experiment. However, it has been shown that the percentage of overlapping genes is of limited use for comparing two experiments due to the variability of gene signatures caused by different array platforms or assay-specific influencing parameters. Here, we present a robust approach for a systematic and quantitative comparison of published gene expression signatures with an exemplary query dataset.