Selection bias in gene extraction on the basis of microarray gene-expression data

In the context of cancer diagnosis and treatment, we consider the problem of constructing an accurate prediction rule on the basis of a relatively small number of tumor tissue samples of known type containing the expression data on very many (possibly thousands) genes. Recently, results have been presented in the literature suggesting that it is possible to construct a prediction rule from only a few genes such that it has a negligible prediction error rate. However, in these results the test error or the leave-one-out cross-validated error is calculated without allowance for the selection bias. There is no allowance because the rule is either tested on tissue samples that were used in the first instance to select the genes being used in the rule or because the cross-validation of the rule is not external to the selection process; that is, gene selection is not performed in training the rule at each stage of the cross-validation process. We describe how in practice the selection bias can be assessed and corrected for by either performing a cross-validation or applying the bootstrap external to the selection process. We recommend using 10-fold rather than leave-one-out cross-validation, and concerning the bootstrap, we suggest using the so-called. 632+ bootstrap error estimate designed to handle overfitted prediction rules. Using two published data sets, we demonstrate that when correction is made for the selection bias, the cross-validated error is no longer zero for a subset of only a few genes.

Modelling High-Dimensional Data by Mixtures of Factor Analyzers

We focus on mixtures of factor analyzers from the perspective of a method for model-based density estimation from high-dimensional data, and hence for the clustering of such data. This approach enables a normal mixture model to be fitted to a sample of n data points of dimension p, where p is large relative to n. The number of free parameters is controlled through the dimension of the latent factor space. By working in this reduced space, it allows a model for each component-covariance matrix with complexity lying between that of the isotropic and full covariance structure models. We shall illustrate the use of mixtures of factor analyzers in a practical example that considers the clustering of cell lines on the basis of gene expressions from microarray experiments. (C) 2002 Elsevier Science B.V. All rights reserved.

Clustering Process Analysis in a Large-Size Dropshaft and in a Hydraulic Jump

Recent efforts in the characterization of air-water flows properties have included some clustering process analysis. A cluster of bubbles is defined as a group of two or more bubbles, with a distinct separation from other bubbles before and after the cluster. The present paper compares the results of clustering processes two hydraulic structures. That is, a large-size dropshaft and a hydraulic jump in a rectangular horizontal channel. The comparison highlighted some significant differences in clustering production and structures. Both dropshaft and hydraulic jump flows are complex turbulent shear flows, and some clustering index may provide some measure of the bubble-turbulence interactions and associated energy dissipation.

Supervisory control of wastewater treatment plants by combining principal component analysis and fuzzy c-means clustering

In this paper a methodology for integrated multivariate monitoring and control of biological wastewater treatment plants during extreme events is presented. To monitor the process, on-line dynamic principal component analysis (PCA) is performed on the process data to extract the principal components that represent the underlying mechanisms of the process. Fuzzy c-means (FCM) clustering is used to classify the operational state. Performing clustering on scores from PCA solves computational problems as well as increases robustness due to noise attenuation. The class-membership information from FCM is used to derive adequate control set points for the local control loops. The methodology is illustrated by a simulation study of a biological wastewater treatment plant, on which disturbances of various types are imposed. The results show that the methodology can be used to determine and co-ordinate control actions in order to shift the control objective and improve the effluent quality.

On the statistical analysis of the GS-NS0 cell proteome: Imputation, clustering and variability testing

We have undertaken two-dimensional gel electrophoresis proteomic profiling on a series of cell lines with different recombinant antibody production rates. Due to the nature of gel-based experiments not all protein spots are detected across all samples in an experiment, and hence datasets are invariably incomplete. New approaches are therefore required for the analysis of such graduated datasets. We approached this problem in two ways. Firstly, we applied a missing value imputation technique to calculate missing data points. Secondly, we combined a singular value decomposition based hierarchical clustering with the expression variability test to identify protein spots whose expression correlates with increased antibody production. The results have shown that while imputation of missing data was a useful method to improve the statistical analysis of such data sets, this was of limited use in differentiating between the samples investigated, and highlighted a small number of candidate proteins for further investigation. (c) 2006 Elsevier B.V. All rights reserved.

Equational Reasoning as a Tool for Data Analysis

A combination of deductive reasoning, clustering, and inductive learning is given as an example of a hybrid system for exploratory data analysis. Visualization is replaced by a dialogue with the data.

Sequence analysis of rabbit haemorrhagic disease virus (RHDV) in Australia: alterations after its release

Liver samples from rabbits killed by RHDV, collected from five States in Australia in 1996 and 1997 were analysed by RT-PCR. A 398 bp fragment of the capsid protein (VP60) gene was amplified by PCR and directly sequenced. The alignment of the nucleotide and amino acid sequences and their comparison with the original strain of the virus released in Australia indicated genetic changes after two years have been small with 98.2% to 100% identity. The constructed phylogenetic tree suggests slight differences in nucleotide substitutions in various States but there is no clear evidence of clustering of sequences according to their geographic origin. In practical terms, sequencing of viral RNA provides a means of testing the efficacy of further releases and subsequent spread of the virus if such a strategy is employed as a means of enhancing RHD as a biological control of the wild rabbit in Australia.

A mixture model-based approach to the clustering of microarray expression data

Motivation: This paper introduces the software EMMIX-GENE that has been developed for the specific purpose of a model-based approach to the clustering of microarray expression data, in particular, of tissue samples on a very large number of genes. The latter is a nonstandard problem in parametric cluster analysis because the dimension of the feature space (the number of genes) is typically much greater than the number of tissues. A feasible approach is provided by first selecting a subset of the genes relevant for the clustering of the tissue samples by fitting mixtures of t distributions to rank the genes in order of increasing size of the likelihood ratio statistic for the test of one versus two components in the mixture model. The imposition of a threshold on the likelihood ratio statistic used in conjunction with a threshold on the size of a cluster allows the selection of a relevant set of genes. However, even this reduced set of genes will usually be too large for a normal mixture model to be fitted directly to the tissues, and so the use of mixtures of factor analyzers is exploited to reduce effectively the dimension of the feature space of genes. Results: The usefulness of the EMMIX-GENE approach for the clustering of tissue samples is demonstrated on two well-known data sets on colon and leukaemia tissues. For both data sets, relevant subsets of the genes are able to be selected that reveal interesting clusterings of the tissues that are either consistent with the external classification of the tissues or with background and biological knowledge of these sets.

Model-based clustering in gene expression microarrays: An application to breast cancer data

In microarray studies, the application of clustering techniques is often used to derive meaningful insights into the data. In the past, hierarchical methods have been the primary clustering tool employed to perform this task. The hierarchical algorithms have been mainly applied heuristically to these cluster analysis problems. Further, a major limitation of these methods is their inability to determine the number of clusters. Thus there is a need for a model-based approach to these. clustering problems. To this end, McLachlan et al. [7] developed a mixture model-based algorithm (EMMIX-GENE) for the clustering of tissue samples. To further investigate the EMMIX-GENE procedure as a model-based -approach, we present a case study involving the application of EMMIX-GENE to the breast cancer data as studied recently in van 't Veer et al. [10]. Our analysis considers the problem of clustering the tissue samples on the basis of the genes which is a non-standard problem because the number of genes greatly exceed the number of tissue samples. We demonstrate how EMMIX-GENE can be useful in reducing the initial set of genes down to a more computationally manageable size. The results from this analysis also emphasise the difficulty associated with the task of separating two tissue groups on the basis of a particular subset of genes. These results also shed light on why supervised methods have such a high misallocation error rate for the breast cancer data.

Document retrieval for e-mail search and discovery using Formal Concept Analysis

This paper discusses a document discovery tool based on Conceptual Clustering by Formal Concept Analysis. The program allows users to navigate e-mail using a visual lattice metaphor rather than a tree. It implements a virtual. le structure over e-mail where files and entire directories can appear in multiple positions. The content and shape of the lattice formed by the conceptual ontology can assist in e-mail discovery. The system described provides more flexibility in retrieving stored e-mails than what is normally available in e-mail clients. The paper discusses how conceptual ontologies can leverage traditional document retrieval systems and aid knowledge discovery in document collections.

Mixture modelling for cluster analysis

Cluster analysis via a finite mixture model approach is considered. With this approach to clustering, the data can be partitioned into a specified number of clusters g by first fitting a mixture model with g components. An outright clustering of the data is then obtained by assigning an observation to the component to which it has the highest estimated posterior probability of belonging; that is, the ith cluster consists of those observations assigned to the ith component (i = 1,..., g). The focus is on the use of mixtures of normal components for the cluster analysis of data that can be regarded as being continuous. But attention is also given to the case of mixed data, where the observations consist of both continuous and discrete variables.

Construction of representative transcript and protein sets of human, mouse, and rat as a platform for their transcriptome and proteome analysis

The number of mammalian transcripts identified by full-length cDNA projects and genome sequencing projects is increasing remarkably. Clustering them into a strictly nonredundant and comprehensive set provides a platform for functional analysis of the transcriptome and proteome, but the quality of the clustering and predictive usefulness have previously required manual curation to identify truncated transcripts and inappropriate clustering of closely related sequences. A Representative Transcript and Protein Sets (RTPS) pipeline was previously designed to identify the nonredundant and comprehensive set of mouse transcripts based on clustering of a large mouse full-length cDNA set (FANTOM2). Here we propose an alternative method that is more robust, requires less manual curation, and is applicable to other organisms in addition to mouse. RTPSs of human, mouse, and rat have been produced by this method and used for validation. Their comprehensiveness and quality are discussed by comparison with other clustering approaches. The RTPSs are available at ftp://fantom2.gsc.riken.go.jp/RTPS/. (C). 2004 Elsevier Inc. All rights reserved.

Cluster analysis of high-dimensional data: A case study

Normal mixture models are often used to cluster continuous data. However, conventional approaches for fitting these models will have problems in producing nonsingular estimates of the component-covariance matrices when the dimension of the observations is large relative to the number of observations. In this case, methods such as principal components analysis (PCA) and the mixture of factor analyzers model can be adopted to avoid these estimation problems. We examine these approaches applied to the Cabernet wine data set of Ashenfelter (1999), considering the clustering of both the wines and the judges, and comparing our results with another analysis. The mixture of factor analyzers model proves particularly effective in clustering the wines, accurately classifying many of the wines by location.

Socioeconomic position, cognitive function, and clustering of cardiovascular risk factors in adolescence: Findings from the Mater University Study of Pregnancy and its outcomes

Objectives: The objectives of this study were to examine the extent of clustering of smoking, high levels of television watching, overweight, and high blood pressure among adolescents and whether this clustering varies by socioeconomic position and Cognitive function. Methods: This study was a cross-sectional analysis of 3613 (1742 females) participants of an Australian birth cohort who were examined at age 14. Results: Three hundred fifty-three (9.8%) of the participants had co-occurrence of three or four risk factors. Risk factors clustered in these adolescents with a greater number of participants than would be predicted by assumptions of independence having no risk factors and three or four risk factors. The extent of clustering tended to be greater in those from lower-income families and among those with lower cognitive function. The age-adjusted ratio of observed to expected cooccurrence of three or four risk factors was 2.70 (95% confidence interval [Cl], 1.80-4.06) among those from low-income families and 1.70 (95% Cl, 1.34-2.16) among those from more affluent families. The ratio among those with low Raven's scores (nonverbal reasoning) was 2.36 (95% Cl, 1.69-3.30) and among those with higher scores was 1.51 (95% Cl, 1.19-1.92); similar results for the WRAT 3 score (reading ability) were 2.69 (95% Cl, 1.85-3.94) and 1.68 (95% Cl, 1.34-2.11). Clustering did not differ by sex. Conclusion: Among adolescents, coronary heart disease risk factors cluster, and there is some evidence that this clustering is greater among those from families with low income and those who have lower cognitive function.