No description available.

Correspondence analysis is a descriptive and explorative technique for the study of associations between variables. It is a visualization method for analyzing high dimensional data via projections onto a low-dimensional subspace. In this thesis, we demonstrate the applicability of correspondence analysis to microarray data. We show that it can be used to identify important genes and treatment patterns by coordinating and projecting the genes and the experimental conditions. In addition, we estimate missing values in the gene expressions using the Expectation-Maximization (EM) algorithm and identify genes with large between-condition variability using the projections of the genes and the conditions. To demonstrate its application, correspondence analysis is applied to various simulated data and microarray data from the EPA (Environmental Protection Agency) studies. We conclude that correspondence analysis is a useful tool for analyzing the associations between genes and experimental conditions, for identifying important genes, and for estimating missing values.

Title: Multidimensional statistics and applications to study genes Author: Mgr. Peter Bubelíny Department: Department of probability and mathematical statistics Supervisor: prof. Lev Klebanov, DrSc., KPMS MFF UK Abstract: Microarray data of gene expressions consist of thousands of genes and just some tens of observations. Moreover, genes are highly correlated between themselves and contain systematic errors. Hence the magnitude of these data does not afford us to estimate their correlation structure. In many statistical problems with microarray data, we have to test some thousands of hypotheses simultaneously. Due to dependence between genes, p-values of these hypotheses are dependent as well. In this work, we compared conve- nient multiple testing procedures reasonable for dependent hypotheses. The common manner to make microarray data more uncorrelated and partially eliminate systematic errors is normalizing them. We proposed some new normalizations and studied how different normalizations influence hypothe- ses testing. Moreover, we compared tests for finding differentially expressed genes or gene sets and identified some interesting properties of some tests such as bias of two-sample Kolmogorov-Smirnov test and interesting behav- ior of Hotelling's test for dependent components of observations. In the end of...

Sorghum crop has become attractive to breeders due to its drought tolerance, and many uses including a human food source, animal feed, industrial fibre and bioenergy crop. Sorghum, like any other plant, is a host to a variety of microbes that can have neutral, negative or positive effects on the plant. While the majority of microorganisms are beneficial, pathogens colonize plant tissues and overwhelm its defence mechanisms. This colonization is a direct threat to the sorghum productivity. The development of microbiome-based approaches for sustainable crop productivity and yield is hindered by a lack of understanding of the main biotic factors affecting the crop microbiome. Metabarcoding has proven to be a valuable tool which has been widely used for characterizing the microbial diversity and composition of different environments and has been utilized in many research endeavours. This study analyses the relationship between the microbiota and their response to natural pathogen infection in sorghum disease groups (R, MR, S and HS) and identifies the most dominant pathogen in the highly susceptible disease group. The study also, assesses the spore viability through the use of the automated cell counter and confirms Fusarium graminearum (dominant pathogen linked to the HS disease group) through sequencing of the marker genes, to subsequently characterize pathways likely to be involved in pathogen infection resistance. To achieve the objectives, a combination of 16S rRNA (V3/V4 regions) and ITS (ITS1/ITS4) of the internal transcribed spacer regions were amplified and sequenced using NGS technologies to study the microbiota in response to natural infection. Additionally, comparative transcriptional analysis of sorghum RILs in response to Fusarium graminearum infection was conducted through RNA-Seq. Upon natural infection, the foliar symptoms assessment of the RILs was conducted and four disease groups; resistant (R), moderately resistant (MR), susceptible (S) and highly susceptible (HS) were designated. The results of the present metabarcoding study indicate that resistant sorghum leaves (R group) supported a large diversity of fungal and bacterial microbes. The genera Methylorubrum, Enterobacter and Sphingomonas with reported plant growth promoting traits were more abundant and highly enriched in the R and MR group, with members of the latter genus significantly enriched in the R group. The resistant fungal group had a majority of OTUs showing similarity to well-known plant growth-promoting fungal genus including Papiliotrema (Tremellaceae family), which are known biocontrol agents. The yeast Hannaella was also highly linked with the resistant plants. Some Hannaella species are known to produce indole acetic acid (IAA) for promoting plant growth. Metabarcoding was also used to assess the major potential disease-causing taxa associated with the highly diseased group. It identified fungal pathogenic species, that have not previously been identified as pathogens of sorghum such as Ascochyta paspali and Ustilago kamerunensis (which are known pathogenic fungi of grass species) and were associated with the susceptible disease groups (S and HS). These analyses revealed the potential sorghum fungal pathogen Epicoccum sorghinum, and was highly linked with the S disease group. It further expanded the identification of a reportedly economically importance species causing sorghum related diseases Fusarium graminearum (anamorph Gibberella zeae). This species has also been identified in this study to be highly associated with the RILs showing major disease symptoms. Fusarium graminearum a significant pathogen in winter cereals and maize has been associated with stalk rot of sorghum and sorghum grain mould. The presence of Fusarium graminearum in sorghum can be a toxicological risk, since this species has the potential to produce mycotoxins. It was further shown that natural pathogen infection results in distinct foliar microbial communities in sorghum RILs. The co-occurrence taxa represented by Tremellomycetes and Dothiomycetes fungal classes and Bacillaceae and Sphingomonadaceae bacterial family had more central roles in the network. The modules which are located centrally on the network have been expected to play important ‘topological roles’ in interconnecting pairs of other fungal and bacterial taxa in the symbiont–symbiont co-occurrence network. These taxa having a central role, are considered to be keystone microbes, and have been suggested to be drivers of microbiome structure and functioning. The results of bacterial and fungal community composition, community co-occurrences further suggested the importance of keystone taxa which may disproportionately shape the structure of foliar microbiomes. The foliar disease symptom assessments revealed that sorghum RIL 131 was highly diseased and RIL 103 did not show any visible disease symptoms and were subsequently used for transcriptomic analysis. Gene expression patterns were studied between the identified RIL that did not show visible symptoms (resistant RIL no 103) and the RIL that showed major disease symptoms (susceptible RIL no 131). Fusarium graminearum the dominant potential pathogen found in this study to be associated with the highly susceptible plants was used to inoculate RILs at seedling stage in a greenhouse and samples were collected in triplicates at 24 hours post infection (hpi), 48 hpi, 7 days post infection (dpi) and 14 dpi. Prior to that, ITS and UBC genes confirmed the identity of Fusarium graminearum, and the automated haemocytometer confirmed the cell/spore viability. Using RNA-Seq analysis it was shown that the resistant RIL had defence related pathways from early response (24- 48 hpi) to late response (7-14 dpi). And the more the infection progressed, the more the defence related genes were up-regulated in terms of fragments per kilobase of exon model per million reads mapped (FPKM) and False Discovery Rate (FDR ≤ 0.05) values. Transcriptome time series expression profiling was used to characterize the plant response to Fusarium graminearum with the Dirichlet Process Gaussian Process mixture model software (DPGP) in susceptible and resistant RILs. The susceptible RIL (number 131) transcriptional response upon Fusarium graminearum infection presented differences of the closely related clustered expression profiles across all timepoints in both RILs. Group 2 exclusively clustered the genes encoding the sesquiterpene metabolism pathway, which is one of the major physiological change occurring in response to fungal infection and has been previously reported to produce the mycotoxins associated with Fusarium head blight (FHB) of cereals. This pathway presented an increase from the initial infection phase to the late infection phase in group 4, the genes encoding starch sucrose, metabolism and cyanoamino acid pathways presented a pattern that had a sharp decline from 48 hpi -14 dpi (at a later stage of infection). This could suggest that, as the time progresses in the susceptible RIL the pathways which are important in plant defence declines at a late infection stage. Group 3 presented a pattern increase of the 5-lipoxygenase (LOX 5) gene expressed from 48 hpi-14 dpi timepoints. The loss and silencing of LOX5 function have in the past described to be linked with enhanced disease resistance. In this study the LOX5 was expressed and this could suggest that LOX5 might have a function as a susceptibility factor in disease caused by Fusarium graminearum in sorghum RILs. CBL-interacting protein kinase 6 (CIPK6) gene was also associated with this group. This gene has been associated with negative regulation of immune response to Pseudomonas syringae in Arabidopsis as plants overexpressing CIPK6 were more susceptible to Pseudomonas syringae. Transcriptional response of a resistant RIL (number 103) to infection with Fusarium graminearum presented an increase in genes encoding metabolic and biosynthesis of metabolites pathways in group 1 and group 4 at early infection phase and a sharp decline in the late infection phase. An increase in the genes encoding pathways in earlier infection state could suggest the establishment of a beneficial energy balance for defence. Additionally, genes encoding phenylpropanoid (PAL), galactose and glycolysis pathway were amongst the genes increased at early stages of infection in group 1. Sugar can play a significant role in resistance to fungal pathogens through phenylpropanoid metabolism stimulation, and previous studies showed that the phenylpropanoid pathway could play a role in resistance of wheat to Fusarium graminearum and deoxynivalenol. Overall, this study represents a first step in understanding the molecular mechanisms involved in resistance to Fusarium graminearum. This analysis has also identified the reported beneficial microbes and defence related genes and pathways. Together, the current findings suggest that different ‘resident’ consortia found in naturally infected and uninfected sorghum plants may be viable biocontrol and plant-growth promoting targets. Cultivation studies may shed light on the nature of the putative symbiotic relationships between bacteria and fungi. These results have consequences for crop breeding, and the analysis of microbial diversity and community composition can be useful biomarkers for assessing disease status in plants. The transcriptome and metabarcoding data generated will help guide further research to develop novel strategies for management of disease in sorghum RILs through the integrative approach considering both beneficial microbes and defence related genes. This provides the baseline information and will positively impact in the development of Fusarium graminearum resistant genotypes in future through the integration/incorporation of beneficial microorganisms (bacteria and fungi) and resistant genes in breeding strategies. / Life and Consumer Sciences / D. Phil. (Life Sciences)

Yes / Cytokines are extremely potent biomolecules that regulate cellular functions and play multiple roles in initiation and inhibition of disease. These highly specialised macromolecules are actively involved in control of cellular proliferation, apoptosis, cell migration and adhesion. This work, investigates the effect of transforming growth factor-beta2 (TGF-β2) on the biological regulation of chondrocyte and the repair of a created model wound on a multilayer culture system. Also the effect of this cytokine on cell length, proliferation, and cell adhesion has been investigated. Chondrocytes isolated from knee joint of rats and cultured at 4 layers. Each layer consisted of 2 × 105 cells/ml with and without TGF-β2. The expression of mRNA and protein levels of TGF-β receptors and Smad1, 3, 4, and 7 have been analysed by RT-PCR and western blot analysis. The effect of different supplementations in chondrocyte cell proliferation, cell length, adhesion, and wound repair was statistically analysed by One-way ANOVA test. Our results showed that the TGFβ2 regulates mRNA levels of its own receptors, and of Smad3 and Smad7. Also the TGF-β2 caused an increase in chondrocyte cell length, but decreased its proliferation rate and the wound healing process. TGF-β2 also decreased cell adhesion ability to the surface of the culture flask. Since, TGF-β2 increased the cell size, but showed negative effect on cell proliferation and adhesion of CHC, the effect of manipulated TGF-β2 with other growth factors and/or proteins needs to be investigated to finalize the utilization of this growth factor and design of scaffolding in treatment of different types of arthritis.

In this dissertation, we propose and develop innovative data modeling and analysis methods for extracting meaningful and specific information about disease mechanisms from microarray gene expression data. To provide a high-level overview of gene expression data for easy and insightful understanding of data structure, we propose a novel statistical data clustering and visualization algorithm that is comprehensively effective for multiple clustering tasks and that overcomes some major limitations of existing clustering methods. The proposed clustering and visualization algorithm performs progressive, divisive hierarchical clustering and visualization, supported by hierarchical statistical modeling, supervised/unsupervised informative gene/feature selection, supervised/unsupervised data visualization, and user/prior knowledge guidance through human-data interactions, to discover cluster structure within complex, high-dimensional gene expression data. For the purpose of selecting suitable clustering algorithm(s) for gene expression data analysis, we design an objective and reliable clustering evaluation scheme to assess the performance of clustering algorithms by comparing their sample clustering outcome to phenotype categories. Using the proposed evaluation scheme, we compared the performance of our newly developed clustering algorithm with those of several benchmark clustering methods, and demonstrated the superior and stable performance of the proposed clustering algorithm. To identify the underlying active biological processes that jointly form the observed biological event, we propose a latent linear mixture model that quantitatively describes how the observed gene expressions are generated by a process of mixing the latent active biological processes. We prove a series of theorems to show the identifiability of the noise-free model. Based on relevant geometric concepts, convex analysis and optimization, gene clustering, and model stability analysis, we develop a robust blind source separation method that fits the model to the gene expression data and subsequently identify the underlying biological processes and their activity levels under different biological conditions. Based on the experimental results obtained on cancer, muscle regeneration, and muscular dystrophy gene expression data, we believe that the research work presented in this dissertation not only contributes to the engineering research areas of machine learning and pattern recognition, but also provides novel and effective solutions to potentially solve many biomedical research problems, for improving the understanding about disease mechanisms. / Ph. D.

DNA microarray has become a powerful tool in genetics, molecular biology, and biomedical research. DNA microarray can be used for measuring the genotypes, structural changes, and gene expressions of human genomes. Detection and characterization of multilevel, high-throughput microarray genomic data pose new challenges to statistical pattern recognition and machine learning research. In this dissertation, we propose novel computational methods for analyzing DNA copy number changes and learning the trees of phenotypes using DNA microarray data. DNA copy number change is an important form of structural variations in human genomes. The copy number signals measured by high-density DNA microarrays usually have low signal-to-noise ratios and complex patterns due to inhomogeneous composition of tissue samples. We propose a robust detection method for extracting copy number changes in a single signal profile and consensus copy number changes in the signal profiles of a population. We adapt a solution-path algorithm to efficiently solve the optimization problems associated with the proposed method. We tested the proposed method on both simulation and real CGH and SNP microarray datasets, and observed competitively improved performance as compared to several widely-adopted copy number change detection methods. We also propose a chromosome instability measure to summarize the extracted copy number changes for assessing chromosomal instabilities of tumor genomes. The proposed measure demonstrates distinct patterns between different subtypes of ovarian serous carcinomas and normal samples. Among active research on complex human diseases using genomic data, little effort and progress have been made in discovering the relational structural information embedded in the molecular data. We propose two stability analysis based methods to learn stable and highly resolved trees of phenotypes using microarray gene expression data of heterogeneous diseases. In the first method, we use a hierarchical, divisive visualization approach to explore the tree of phenotypes and a leave-one-out cross validation to select stable tree structures. In the second method, we propose a node bandwidth constraint to construct stable trees that can balance the descriptive power and reproducibility of tree structures. Using a top-down merging procedure, we modify the binary tree structures learned by hierarchical group clustering methods to achieve a given node bandwidth. We use a bootstrap based stability analysis to select stable tree structures under different node bandwidth constraints. The experimental results on two microarray gene expression datasets of human diseases show that the proposed methods can discover stable trees of phenotypes that reveal the relationships between multiple diseases with biological plausibility. / Ph. D.

Increased oxidative stress and impaired antioxidant defense mechanisms are believed to be the important factors contributing to the pathogenesis and progression of diabetes mellitus. The products of lipid peroxidation and protein oxidation reactions were all found to be elevated significantly (p&lt / 0.05) in diabetic animals and supplementing the animals either individually or in combination, with two powerful antioxidants DL-&amp / #945 / -lipoic acid (LA) and vitamin C (VC) brought this increment toward the control values. Considering Cu-Zn SOD, CAT and GST-Mu, there was a significant decrease in all activities in diabetic group as compared with control animals. RT-PCR and Western blot analysis results demonstrated that this decrease in activity is regulated at the level of gene expression, as both mRNA and protein expressions were also suppressed for these enzymes. However, in diabetic animals both the mRNA expressions and the activities of two other antioxidant enzymes, namely Mn SOD and GPx, did not change, indicating that the control of activities of these two enzymes were not at the level of genes. Supplementing the diabetic animals with VC increased all CAT, Cu-Zn SOD, GPx, and GST-Mu activities without changing both mRNA and protein expressions suggesting the possible role of post-translational modifications. On the other hand, the effect of VC on Mn SOD was observed at mRNA levels reflecting a transcriptional regulation. Furthermore, supplementing the animals with LA increased the CAT, Cu-Zn SOD, Mn SOD and GPx activities in diabetic rats but different from VC, LA also increased mRNA of CAT and protein levels of CAT, Cu-Zn SOD and Mn SOD suggesting both transcriptional and translational regulation showed by LA. Combined application of antioxidants also increased the CAT, Cu-Zn SOD, Mn SOD and GPx activities toward the control values, but this time there were no statistically significant change in their mRNA expressions even though protein amounts of both CAT and GPx were augmented. That is, when given together, these antioxidants exert their effects mainly at the level of protein synthesis. As a conclusion, diabetes and the resulting oxidative stress coordinately regulate the activities of the antioxidant enzymes at different regulatory points. LA and VC, two powerful antioxidants affect all antioxidant enzyme activities at different levels of transcription and translation. The results indicated the presence of very intricate control mechanisms regulating the activities of antioxidant enzymes in order to prevent the damaging effects of oxidative stress.

Le traitement au froid en post-récolte (TPR) des ananas (Ananas comosus. (L). M) destiné à ralentir la sénescence des fruits, induit « le brunissement interne de l'ananas » (BI). Afin d'étudier des gènes susceptibles de discriminer les variétés tolérantes, une comparaison entre des fruits frais ou soumis au TPR a été réalisée pour 4 variétés d'ananas différant par leur tolérance au BI. D'après les résultats, en réponse au TPR l'absence de symptômes de BI est associée à une « tolérance membranaire » et à une faible activité de la polyphenol oxydase. L'étude de l'activité de la phenylalanine ammonia-lyase et de l'ascorbate peroxidase a révélé que le froid induit une stimulation de l'activité de ces enzymes chez une variété sensible au BI. L'étude du catabolisme protéique a montré que la tolérance des fruits au BI était liée : à la sous-expression du gène d'une protéase à cystéine, et à une sur-expression de gènes codant une cystatine et une protéase à acide aspartique, dont l'ADNc a été caractérisé et cloné pour la première fois chez l'ananas. L'expression différentielle de ces gènes indique qu'ils pourraient être utilisés pour le criblage par PCR de variétés dans les programmes d'améliorations de l'ananas pour la résistance au BI / Pineapple fruits (Ananas comosus. (L). M) require postharvest chilling treatment (PCT) in order to extend the postharvest fruit quality during shipping exportation. However PCT induces an injury known as blackheart (BH), or fruit browning, which is characterized by the appearance of brown spots in the flesh. This work has focused on the study of the development of BH physiopathy in the context of postharvest treatment in 4 pineapple varieties differing in their resistance to BH. Results showed that BH was associated with high membrane tolerance, low activity of polyphenol oxidase and absence of the related isoforms. Under chilling stress, the activities of both phenylalanine ammonia-lyase and ascorbate peroxidase were enhanced in the BH susceptible variety. Various genes involved in protein catabolism under abiotic stress were also studied. BH resistance was shown to be linked to the down-regulation of a major cystein protease and to the up-regulation of cystatin, the natural inhibitor of cystein protease. An aspartic acid protease, isolated and sequenced for the first time in pineapple, was also studied. Opposed to cystein protease, the expression and activity of the aspartic acid protease was directly related to BH resistance. Taken together, the results gathered by this work suggest that these genes could provide useful molecular markers for PCR variety screening in breeding programs aimed at improving pineapple BH resistance