Statistical analysis and meta-analysis of microarray data

The microarray technology provides a high-throughput technique to study gene expression. Microarrays can help us diagnose different types of cancers, understand biological processes, assess host responses to drugs and pathogens, find markers for specific diseases, and much more. Microarray experiments generate large amounts of data. Thus, effective data processing and analysis are critical for making reliable inferences from the data. ^ The first part of dissertation addresses the problem of finding an optimal set of genes (biomarkers) to classify a set of samples as diseased or normal. Three statistical gene selection methods (GS, GS-NR, and GS-PCA) were developed to identify a set of genes that best differentiate between samples. A comparative study on different classification tools was performed and the best combinations of gene selection and classifiers for multi-class cancer classification were identified. For most of the benchmarking cancer data sets, the gene selection method proposed in this dissertation, GS, outperformed other gene selection methods. The classifiers based on Random Forests, neural network ensembles, and K-nearest neighbor (KNN) showed consistently god performance. A striking commonality among these classifiers is that they all use a committee-based approach, suggesting that ensemble classification methods are superior. ^ The same biological problem may be studied at different research labs and/or performed using different lab protocols or samples. In such situations, it is important to combine results from these efforts. The second part of the dissertation addresses the problem of pooling the results from different independent experiments to obtain improved results. Four statistical pooling techniques (Fisher inverse chi-square method, Logit method. Stouffer's Z transform method, and Liptak-Stouffer weighted Z-method) were investigated in this dissertation. These pooling techniques were applied to the problem of identifying cell cycle-regulated genes in two different yeast species. As a result, improved sets of cell cycle-regulated genes were identified. The last part of dissertation explores the effectiveness of wavelet data transforms for the task of clustering. Discrete wavelet transforms, with an appropriate choice of wavelet bases, were shown to be effective in producing clusters that were biologically more meaningful. ^

Gene Expression Profiles in Radiation Workers Occupationally Exposed to Ionizing Radiation

Ionizing radiation OR) imposes risks to human health and the environment. IR at low doses and low (lose rates has the potency to initiate carcinogenesis. Genotoxic environmental agents such as IR trigger a cascade of signal transduction pathways for cellular protection. In this study, using cDNA microarray technique, we monitored the gene expression profiles in lymphocytes derived from radiation-ex posed individuals (radiation workers). Physical dosimetry records on these patients indicated that the absorbed dose ranged from 0.696 to 39.088 mSv. Gene expression analysis revealed statistically significant transcriptional changes in a total of 78 genes (21 up-regulated and 57 clown-regulated) involved in several biological processes such as ubiquitin cycle (UHRF2 and PIAS1), DNA repair (LIG3, XPA, ERCC5, RAD52, DCLRE1C), cell cycle regulation/proliferation (RHOA, CABLES2, TGFB2, IL16), and stress response (GSTP1, PPP2R5A, DUSP22). Some of the genes that showed altered expression profiles in this study call be used as biomarkers for monitoring the chronic low level exposure in humans. Additionally, alterations in gene expression patterns observed in chronically exposed radiation workers reinforces the need for defining the effective radiation dose that causes immediate genetic damage as well as the long-term effects on genomic instability, including cancer.

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.

Gene expression databases and data mining.

The DNA microarray technology has arguably caught the attention of the worldwide life science community and is now systematically supporting major discoveries in many fields of study. The majority of the initial technical challenges of conducting experiments are being resolved, only to be replaced with new informatics hurdles, including statistical analysis, data visualization, interpretation, and storage. Two systems of databases, one containing expression data and one containing annotation data are quickly becoming essential knowledge repositories of the research community. This present paper surveys several databases, which are considered "pillars" of research and important nodes in the network. This paper focuses on a generalized workflow scheme typical for microarray experiments using two examples related to cancer research. The workflow is used to reference appropriate databases and tools for each step in the process of array experimentation. Additionally, benefits and drawbacks of current array databases are addressed, and suggestions are made for their improvement.

Modular analysis of gene expression data with R.

SUMMARY: Large sets of data, such as expression profiles from many samples, require analytic tools to reduce their complexity. The Iterative Signature Algorithm (ISA) is a biclustering algorithm. It was designed to decompose a large set of data into so-called 'modules'. In the context of gene expression data, these modules consist of subsets of genes that exhibit a coherent expression profile only over a subset of microarray experiments. Genes and arrays may be attributed to multiple modules and the level of required coherence can be varied resulting in different 'resolutions' of the modular mapping. In this short note, we introduce two BioConductor software packages written in GNU R: The isa2 package includes an optimized implementation of the ISA and the eisa package provides a convenient interface to run the ISA, visualize its output and put the biclusters into biological context. Potential users of these packages are all R and BioConductor users dealing with tabular (e.g. gene expression) data. AVAILABILITY: http://www.unil.ch/cbg/ISA CONTACT: sven.bergmann@unil.ch

Cross-species and cross-platform gene expression studies with the Bioconductor-compliant R package 'annotationTools'.

Background: The variety of DNA microarray formats and datasets presently available offers an unprecedented opportunity to perform insightful comparisons of heterogeneous data. Cross-species studies, in particular, have the power of identifying conserved, functionally important molecular processes. Validation of discoveries can now often be performed in readily available public data which frequently requires cross-platform studies.Cross-platform and cross-species analyses require matching probes on different microarray formats. This can be achieved using the information in microarray annotations and additional molecular biology databases, such as orthology databases. Although annotations and other biological information are stored using modern database models ( e. g. relational), they are very often distributed and shared as tables in text files, i.e. flat file databases. This common flat database format thus provides a simple and robust solution to flexibly integrate various sources of information and a basis for the combined analysis of heterogeneous gene expression profiles.Results: We provide annotationTools, a Bioconductor-compliant R package to annotate microarray experiments and integrate heterogeneous gene expression profiles using annotation and other molecular biology information available as flat file databases. First, annotationTools contains a specialized set of functions for mining this widely used database format in a systematic manner. It thus offers a straightforward solution for annotating microarray experiments. Second, building on these basic functions and relying on the combination of information from several databases, it provides tools to easily perform cross-species analyses of gene expression data.Here, we present two example applications of annotationTools that are of direct relevance for the analysis of heterogeneous gene expression profiles, namely a cross-platform mapping of probes and a cross-species mapping of orthologous probes using different orthology databases. We also show how to perform an explorative comparison of disease-related transcriptional changes in human patients and in a genetic mouse model.Conclusion: The R package annotationTools provides a simple solution to handle microarray annotation and orthology tables, as well as other flat molecular biology databases. Thereby, it allows easy integration and analysis of heterogeneous microarray experiments across different technological platforms or species.

eVOC: a controlled vocabulary for unifying gene expression data.

Expression data contribute significantly to the biological value of the sequenced human genome, providing extensive information about gene structure and the pattern of gene expression. ESTs, together with SAGE libraries and microarray experiment information, provide a broad and rich view of the transcriptome. However, it is difficult to perform large-scale expression mining of the data generated by these diverse experimental approaches. Not only is the data stored in disparate locations, but there is frequent ambiguity in the meaning of terms used to describe the source of the material used in the experiment. Untangling semantic differences between the data provided by different resources is therefore largely reliant on the domain knowledge of a human expert. We present here eVOC, a system which associates labelled target cDNAs for microarray experiments, or cDNA libraries and their associated transcripts with controlled terms in a set of hierarchical vocabularies. eVOC consists of four orthogonal controlled vocabularies suitable for describing the domains of human gene expression data including Anatomical System, Cell Type, Pathology and Developmental Stage. We have curated and annotated 7016 cDNA libraries represented in dbEST, as well as 104 SAGE libraries,with expression information,and provide this as an integrated, public resource that allows the linking of transcripts and libraries with expression terms. Both the vocabularies and the vocabulary-annotated libraries can be retrieved from http://www.sanbi.ac.za/evoc/. Several groups are involved in developing this resource with the aim of unifying transcript expression information.

The direct effects of tacrolimus and cyclosporin A on isolated human islets: A functional, survival and gene expression study.

The use of immunosuppressive drugs in transplanted patients is associated with the development of diabetes, possibly due to β-cell toxicity. To better understand the mechanisms leading to post-transplant diabetes, we investigated the actions of prolonged exposure of isolated human islets to therapeutical levels of tacrolimus (Tac) or cyclosporin A (CsA). Islets were isolated from the pancreas of multiorgan donors by enzymatic digestion and density gradient centrifugation. Functional, survival and molecular studies were then performed after 4 days of incubation with therapeutical concentrations of Tac or  CsA. Glucose-induced insulin secretion was significantly decreased in Tac, but not in CsA exposed islets, which was associated with a reduction of the amount of insulin granules as shown by electron microscopy. The percentage of apoptotic β-cells was higher in Tac than CsA exposed islets. Microarray experiments followed by Gene Set Enrichment Analysis revealed that gene expression was more markedly affected upon Tac treatment. In conclusion, Tac and CsA affect features of beta-cell differently, with several changes occurring at the molecular level.

Algorithmic Techniques in Gene Expression Processing. From Imputation to Visualization

The amount of biological data has grown exponentially in recent decades. Modern biotechnologies, such as microarrays and next-generation sequencing, are capable to produce massive amounts of biomedical data in a single experiment. As the amount of the data is rapidly growing there is an urgent need for reliable computational methods for analyzing and visualizing it. This thesis addresses this need by studying how to efficiently and reliably analyze and visualize high-dimensional data, especially that obtained from gene expression microarray experiments. First, we will study the ways to improve the quality of microarray data by replacing (imputing) the missing data entries with the estimated values for these entries. Missing value imputation is a method which is commonly used to make the original incomplete data complete, thus making it easier to be analyzed with statistical and computational methods. Our novel approach was to use curated external biological information as a guide for the missing value imputation. Secondly, we studied the effect of missing value imputation on the downstream data analysis methods like clustering. We compared multiple recent imputation algorithms against 8 publicly available microarray data sets. It was observed that the missing value imputation indeed is a rational way to improve the quality of biological data. The research revealed differences between the clustering results obtained with different imputation methods. On most data sets, the simple and fast k-NN imputation was good enough, but there were also needs for more advanced imputation methods, such as Bayesian Principal Component Algorithm (BPCA). Finally, we studied the visualization of biological network data. Biological interaction networks are examples of the outcome of multiple biological experiments such as using the gene microarray techniques. Such networks are typically very large and highly connected, thus there is a need for fast algorithms for producing visually pleasant layouts. A computationally efficient way to produce layouts of large biological interaction networks was developed. The algorithm uses multilevel optimization within the regular force directed graph layout algorithm.

Modulation des réactions alloimmunitaires par les cytokines maîtresses IFN-γ et TGF-β

L’injection de cellules immunologiquement compétentes à un hôte histo-incompatible amène une réaction qui peut se traduire par la maladie du greffon-contre-l’hôte (GVHD). La GVHD demeure une barrière importante à une utilisation plus répandue de la greffe allogénique de cellules hématopoïétiques (AHCT), pourtant un traitement efficace pour traiter de nombreuses maladies. Une meilleure compréhension des mécanismes qui sous-tendent cette pathologie pourrait en faciliter le traitement et la prévention. L’Interféron-gamma (IFN-γ) et le Transforming Growth Factor-béta (TGF-β) sont deux cytokines maîtresses de l’immunité impliquées dans la fonction et l’homéostasie des cellules greffées. Nous démontrons chez la souris que l’IFN-γ limite la reconstitution lympho-hématopoïétique de façon dose-dépendante en mobilisant des mécanismes d’apoptose et en inhibant la prolifération cellulaire. Le TGF-β est quant à lui généralement connu comme un immunosuppresseur qui contrôle l’immunité en utilisant plusieurs voies de signalisation. Le rôle relatif de ces voies en AHCT est inconnu. Nous avons étudié une de ces voies en greffant des cellules provenant de donneurs déficients pour le gène SMAD3 (SMAD3-KO), un médiateur central de la voie canonique du TGF-β, à des souris histo-incompatibles. Bien que l’absence de SMAD3 ne cause aucune maladie chez nos souris donneuses, l’injection de cellules SMAD3-KO amène une GVHD du colon sévère chez le receveur. Cette atteinte est caractérisée par une différenciation Th1 et une infiltration massive de granulocytes témoignant d’un rôle central de SMAD3 dans la physiologie des lymphocytes T CD4 et des cellules myéloïdes. Nous avons focalisé ensuite nos efforts sur le rôle de SMAD3 chez les lymphocytes T CD4 en sachant que SMAD3 était actif chez les lymphocytes T CD4 tolérants. Nous avons découvert que SMAD3 était rapidement inactivé après une activation des cellules T, suggérant que l’inactivation de SMAD3 était fonctionnellement importante pour briser l’état de tolérance. Des études de micro-puces d’ADNc nous ont montré que SMAD3 contrôlait en effet l’expression de nombreux transcrits de gènes connus comme étant reliés à la tolérance et/ou à des processus biologiques dont les rôles dans le maintien de la tolérance sont plausibles.

Comparaison des méthodes d'analyse de l'expression différentielle basée sur la dépendance des niveaux d'expression

La technologie des microarrays demeure à ce jour un outil important pour la mesure de l'expression génique. Au-delà de la technologie elle-même, l'analyse des données provenant des microarrays constitue un problème statistique complexe, ce qui explique la myriade de méthodes proposées pour le pré-traitement et en particulier, l'analyse de l'expression différentielle. Toutefois, l'absence de données de calibration ou de méthodologie de comparaison appropriée a empêché l'émergence d'un consensus quant aux méthodes d'analyse optimales. En conséquence, la décision de l'analyste de choisir telle méthode plutôt qu'une autre se fera la plupart du temps de façon subjective, en se basant par exemple sur la facilité d'utilisation, l'accès au logiciel ou la popularité. Ce mémoire présente une approche nouvelle au problème de la comparaison des méthodes d'analyse de l'expression différentielle. Plus de 800 pipelines d'analyse sont appliqués à plus d'une centaine d'expériences sur deux plateformes Affymetrix différentes. La performance de chacun des pipelines est évaluée en calculant le niveau moyen de co-régulation par l'entremise de scores d'enrichissements pour différentes collections de signatures moléculaires. L'approche comparative proposée repose donc sur un ensemble varié de données biologiques pertinentes, ne confond pas la reproductibilité avec l'exactitude et peut facilement être appliquée à de nouvelles méthodes. Parmi les méthodes testées, la supériorité de la sommarisation FARMS et de la statistique de l'expression différentielle TREAT est sans équivoque. De plus, les résultats obtenus quant à la statistique d'expression différentielle corroborent les conclusions d'autres études récentes à propos de l'importance de prendre en compte la grandeur du changement en plus de sa significativité statistique.

Histone H2B-R95A mutant identifies the pheromone pathway that signals cell cycle arrest during rapamycin response

La rapamycine est un immunosuppresseur utilisé pour traiter plusieurs types de maladies dont le cancer du rein. Son fonctionnement par l’inhibition de la voie de Tor mène à des changements dans des processus physiologiques, incluant le cycle cellulaire. Chez Saccharomyces cerevisiae, la rapamycine conduit à une altération rapide et globale de l’expression génique, déclenchant un remodelage de la chromatine. Nous proposons que les modifications des histones peuvent jouer un rôle crucial dans le remodelage de la chromatine en réponse à la rapamycine. Notre objectif principal est d’identifier d’une banque de mutants d’histone les variantes qui vont échouer à répondre à la rapamycine dans une tentative de réaliser une caractérisation des modifications d’histone critiques pour la réponse à cette drogue. Ainsi, nous avons réalisé un criblage d’une banque de mutants d’histone et identifié plusieurs mutants d‘histone dont la résistance à la rapamycine a été altérée. Nous avons caractérisé une de ces variantes d’histone, à savoir H2B, qui porte une substitution de l’alanine en arginine en position 95 (H2B-R95A) et démontré que ce mutant est extrêmement résistant à la rapamycine, et non à d’autres drogues. Des immunoprécipitations ont démontré que H2B-R95A est défectueux pour former un complexe avec Spt16, un facteur essentiel pour la dissociation de H2A et H2B de la chromatine, permetant la réplication et la transcription par les ADN et ARN polymérases, respectivement. Des expériences de ChIP-Chip et de micropuce ont démontré que l’arginine 95 de H2B est requise pour recruter Spt16 afin de permettre l’expression d’une multitude de gènes, dont certains font partie de la voie des phéromones. Des évidences seront présentées pour la première fois démontrant que la rapamycine peut activer la voie des phéromones et qu’une défectuosité dans cette voie cause la résistante à cette drogue.

Pea-barley intercropping and short-term subsequent crop effects across European organic cropping conditions

Grain legumes are known to increase the soil mineral nitrogen (N) content, reduce the infection pressure of soil borne pathogens, and hence enhance subsequent cereals yields. Replicated field experiments were performed throughout W. Europe (Denmark, United Kingdom, France, Germany and Italy) to asses the effect of intercropping pea and barley on the N supply to subsequent wheat in organic cropping systems. Pea and barley were grown either as sole crops at the recommended plant density (P100 and B100, respectively) or in replacement (P50B50) or additive (P100B50) intercropping designs. In the replacement design the total relative plant density is kept constant, while the additive design uses the optimal sole crop density for pea supplementing with 'extra' barley plants. The pea and barley crops were followed by winter wheat with and without N application. Additional experiments in Denmark and the United Kingdom included subsequent spring wheat with grass-clover as catch crops. The experiment was repeated over the three cropping seasons of 2003, 2004 and 2005. Irrespective of sites and intercrop design pea-barley intercropping improved the plant resource utilization (water, light, nutrients) to grain N yield with 25-30% using the Land Equivalent ratio. In terms of absolute quantities, sole cropped pea accumulated more N in the grains as compared to the additive design followed by the replacement design and then sole cropped barley. The post harvest soil mineral N content was unaffected by the preceding crops. Under the following winter wheat, the lowest mineral N content was generally found in early spring. Variation in soil mineral N content under the winter wheat between sites and seasons indicated a greater influence of regional climatic conditions and long-term cropping history than annual preceding crop and residue quality. Just as with the soil mineral N, the subsequent crop response to preceding crop was negligible. Soil N balances showed general negative values in the 2-year period, indicating depletion of N independent of preceding crop and cropping strategy. It is recommended to develop more rotational approaches to determine subsequent crop effects in organic cropping systems, since preceding crop effects, especially when including legumes, can occur over several years of cropping.

Modelling simultaneously water content and dry matter dynamics of wheat grains

A model was devised to describe simultaneously the grain masses of water and dry matter against thermal time during grain filling and maturation of winter wheat. The model accounted for a linear increase in water mass of duration anthesis-m(1) (end of rapid water assimilation phase) and rate a, followed by a more stable water mass until in,, after which water mass declined rapidly at rate e. Grain dry matter was described as a linear increase of rate bgf until a maximum size (maxgf) was attained at m(2).The model was fitted to plot data from weekly samples of grains taken from replicated field experiments investigating effects of grain position (apical or medial), fungicide (five contrasting treatments), sowing date (early or late), cultivar (Malacca or Shamrock) and season (2001/2002 and 2002/2003) on grain filling. The model accounted for between 83 and 99% of the variation ( 2) when fitted to data from individual plots, and between 97 and 99% when fitted to treatment means. Endosperm cell number of grains from early-sown plots in the first season were also counted. Differences in maxgf between grain positions and also between cultivars were mostly the result of effects on bgf and were empirically associated with water mass at nil. Fungicide application controlled S. tritici and powdery mildew infection, delayed flag leaf senescence, increased water mass at m(1) (wm(1)), and also increased m(2), bgf and maxgf. Fungicide effects on water mass were detected before fungicide effects on dry matter, but comparison of the effects of individual fungicide treatments showed no evidence that effects on wm(1), nor on endosperm cell numbers at about m(1), were required for fungicide effects on maxgf, (c) 2005 Elsevier B.V. All rights reserved.

Transcriptomic analysis of rhizobium leguminosarum biovar viciae in symbiosis with host plants pisum sativum and Vicia cracca

Rhizobium leguminosarum bv. viciae forms nitrogen-fixing nodules on several legumes, including pea (Pisum sativum) and vetch (Vicia cracca), and has been widely used as a model to study nodule biochemistry. To understand the complex biochemical and developmental changes undergone by R. leguminosarum bv. viciae during bacteroid development, microarray experiments were first performed with cultured bacteria grown on a variety of carbon substrates (glucose, pyruvate, succinate, inositol, acetate, and acetoacetate) and then compared to bacteroids. Bacteroid metabolism is essentially that of dicarboxylate-grown cells (i.e., induction of dicarboxylate transport, gluconeogenesis and alanine synthesis, and repression of sugar utilization). The decarboxylating arm of the tricarboxylic acid cycle is highly induced, as is gamma-aminobutyrate metabolism, particularly in bacteroids from early (7-day) nodules. To investigate bacteroid development, gene expression in bacteroids was analyzed at 7, 15, and 21 days postinoculation of peas. This revealed that bacterial rRNA isolated from pea, but not vetch, is extensively processed in mature bacteroids. In early development (7 days), there were large changes in the expression of regulators, exported and cell surface molecules, multidrug exporters, and heat and cold shock proteins. fix genes were induced early but continued to increase in mature bacteroids, while nif genes were induced strongly in older bacteroids. Mutation of 37 genes that were strongly upregulated in mature bacteroids revealed that none were essential for nitrogen fixation. However, screening of 3,072 mini-Tn5 mutants on peas revealed previously uncharacterized genes essential for nitrogen fixation. These encoded a potential magnesium transporter, an AAA domain protein, and proteins involved in cytochrome synthesis.