Дисертації з теми "Analisi microarray"

Per varie ragioni, tra cui le loro specifiche condizioni di crescita, la diagnosi di infezione e di contaminazione alimentare da clostridi presenta ancora numerose difficoltà sia a livello clinico-batteriologico che a livello molecolare. In questo lavoro di tesi si è cercato di ampliare lo spettro di conoscenze riguardo i clostridi e la loro diffusione; durante il primo anno di ricerca è stato studiato, applicando nuove tecniche di microscopia, il processo di germinazione di Clostridium tyrobutyricum, uno dei batteri maggiormente responsabili del gonfiore tardivo dei formaggi a pasta dura; l’applicazione di tecniche di Real-Time PCR ha nel contempo reso possibile una determinazione quantitativa dello stesso in latte. Successivamente, è stata condotta un’analisi di tipizzazione molecolare di clostridi nell’ambito di una filiera agro-zoo-casearia finalizzata alle matrici di processo al fine di individuare le possibili vie di diffusione dei microrganismi. La parte finale del lavoro è stata dedicata allo studio di espressione genica di un altro Clostridium responsabile di gonfiore ma scelto perché geneticamente indistinguibile da Clostridium botulinum, ovvero il Clostridium sporogenes; l’analisi trascrizionale dei suoi geni durante le fasi vegetativa, di sporulazione, germinazione ed esocrescita ha permesso di assegnare diverse funzioni a geni singoli o a gruppi di geni allo scopo di utilizzare queste informazioni per formulare ipotesi future anche su altre specie di clostridi patogeni.
For several reasons, including their specific growth requirements, the diagnosis of infections and food contamination caused by clostridia still presents much difficulty at the clinical, bacteriological and molecular levels. The main purpose of this work is to learn more about clostridia and their interactions with environment. First, new microscopy techniques have been used to study the germination process in Clostridium tyrobutyricum, an anaerobic bacterium responsible for late blowing defects during cheese ripening; meanwhile, the application of real-time PCR methods have been employed to enumerate C. tyrobutyricum cells and spores in milk. Then, a molecular genotyping has been set in order to identify the most common clostridia in a agro-dairy production aimed to detect the possible ways of diffusion of these microbial species. The last part concerns the study of expression patterns of Clostridium sporogenes, an apathogenic gram positive clostridium usually involved in food damage and frequently isolated from late bowled cheese; Clostridium sporogenes is genetically indistinguishable from Clostridium botulinum and is often used as a model for the toxic subtypes. The objective of this study is to use an array-based large-scale transcriptional analysis in order to study gene expression in four different steps of Clostridium sporogenes life cycle: vegetative cells, sporulating cells, dormant spores and germinating ones. Our aims includes being able to relate gene-expression patterns to specific phenotypes and to discover gene expression divergences between the different phases of living, germination and outgrowth of spore-forming bacteria. An important aim is to assign functions to groups of or individual C. sporogenes genes and use this information to formulate specific hypotheses for further testing also on pathogenic clostridia types.

Per varie ragioni, tra cui le loro specifiche condizioni di crescita, la diagnosi di infezione e di contaminazione alimentare da clostridi presenta ancora numerose difficoltà sia a livello clinico-batteriologico che a livello molecolare. In questo lavoro di tesi si è cercato di ampliare lo spettro di conoscenze riguardo i clostridi e la loro diffusione; durante il primo anno di ricerca è stato studiato, applicando nuove tecniche di microscopia, il processo di germinazione di Clostridium tyrobutyricum, uno dei batteri maggiormente responsabili del gonfiore tardivo dei formaggi a pasta dura; l’applicazione di tecniche di Real-Time PCR ha nel contempo reso possibile una determinazione quantitativa dello stesso in latte. Successivamente, è stata condotta un’analisi di tipizzazione molecolare di clostridi nell’ambito di una filiera agro-zoo-casearia finalizzata alle matrici di processo al fine di individuare le possibili vie di diffusione dei microrganismi. La parte finale del lavoro è stata dedicata allo studio di espressione genica di un altro Clostridium responsabile di gonfiore ma scelto perché geneticamente indistinguibile da Clostridium botulinum, ovvero il Clostridium sporogenes; l’analisi trascrizionale dei suoi geni durante le fasi vegetativa, di sporulazione, germinazione ed esocrescita ha permesso di assegnare diverse funzioni a geni singoli o a gruppi di geni allo scopo di utilizzare queste informazioni per formulare ipotesi future anche su altre specie di clostridi patogeni.
For several reasons, including their specific growth requirements, the diagnosis of infections and food contamination caused by clostridia still presents much difficulty at the clinical, bacteriological and molecular levels. The main purpose of this work is to learn more about clostridia and their interactions with environment. First, new microscopy techniques have been used to study the germination process in Clostridium tyrobutyricum, an anaerobic bacterium responsible for late blowing defects during cheese ripening; meanwhile, the application of real-time PCR methods have been employed to enumerate C. tyrobutyricum cells and spores in milk. Then, a molecular genotyping has been set in order to identify the most common clostridia in a agro-dairy production aimed to detect the possible ways of diffusion of these microbial species. The last part concerns the study of expression patterns of Clostridium sporogenes, an apathogenic gram positive clostridium usually involved in food damage and frequently isolated from late bowled cheese; Clostridium sporogenes is genetically indistinguishable from Clostridium botulinum and is often used as a model for the toxic subtypes. The objective of this study is to use an array-based large-scale transcriptional analysis in order to study gene expression in four different steps of Clostridium sporogenes life cycle: vegetative cells, sporulating cells, dormant spores and germinating ones. Our aims includes being able to relate gene-expression patterns to specific phenotypes and to discover gene expression divergences between the different phases of living, germination and outgrowth of spore-forming bacteria. An important aim is to assign functions to groups of or individual C. sporogenes genes and use this information to formulate specific hypotheses for further testing also on pathogenic clostridia types.

Il carcinoma epatocellulare (HCC) è la più frequente patologia neoplastica primitiva del fegato, il quinto tumore maligno in ordine di frequenza nel mondo e la terza causa di morte correlata a neoplasia maligna. La sua incidenza annuale è in costante crescita. Tra i fattori patologici che influenzano la sopravvivenza dei pazienti con HCC, la dimensione tumorale, il grado e l’invasione vascolare sono alcuni dei più significativi. Pazienti con tumori di piccole dimensioni (< cm. 3) e senza invasione vascolare hanno una sopravvivenza di circa il 50% in 5 anni. Dato che l’invasione vascolare ed il grado sono criteri difficili da valutare su biopsie, la ricerca si è concentrata sullo studio della biologia dell’HCC nella speranza di individuare marcatori molecolari predittivi del comportamento della malattia. In questo studio sono stati selezionati alcuni biomarcatori (alfa-tubulina, beta-tubulina, LAMA3, osteopontina, Ep-CAM, PAK1), associati a una prognosi sfavorevole ed alcuni marcatori recentemente usati come marcatori diagnostici (glipican3, glutamina sintetasi, heat shock protein 70) al fine di verificare se la loro iperespressione ha un potere predittivo sul comportamento dell’HCC. - Materiali e metodi. Per testare la sensibilità e specificità di questi marcatori, abbiamo usato 1) un microarray tissutale “occidentale” costituito da 98 casi di HCC, HCV-correlati e 2) un microarray tissutale “orientale” costituito da 136 casi di HCC, HBV-correlati. - Risultati. Abbiamo rilevato che le tubuline sono gli unici marcatori che si sono rivelati capaci di fornire informazioni prognostiche e consistenti sulle due popolazioni. OPN, Pak1 e Hsp70, singolarmente o combinati, possono predire una prognosi sfavorevole se applicati alla popolazione orientale. Gli altri marcatori devono essere ancora validati con ulteriori studi prima che si possano definire come marcatori prognostici dell’HCC. - Conclusioni. Infine, 1) abbiamo osservato un’espressione differente dei vari marcatori nelle due casistiche; 2) queste differenze sull’espressione dei marcatori può essere un riflesso delle differenze genetiche, eziologiche e epidemiologiche delle 2 popolazioni; 3) i parametri che emergono all’analisi multivariata sono tuttora criteri patologici: grado e angioinvasione macrovascolare, pertanto i nuovi anticorpi che saranno sviluppati dovranno essere confrontati a questi parametri.
Introduction - HCC ranks among the most lethal cancer in the world with rising incidence. Attempts have been made to predict prognosis in patients with HCC using histopathological features. Tumour grade, size, number of lesions, micro and macrovascular invasion have been correlated with tumor relapse and patient’s survival. However, despite the several therapeutic options today available (tumor ablation, resection, transplantation, chemotherapy and the recent medical therapy with biological drugs) , HCC treatment is largely dictated by gross macroscopic features such as the tumor size and the number of lesions. Thus there is a consistent need to identify tissue biomarkers as individual fingerprints to predict individual. In recent years the interest in molecular biomarkers of HCC genesis and progression has grown, both in terms of prognostic significance and of potential therapeutic targets. We have therefore selected a number of proteins involved in critical cell functions such as staminality, differentiation, adhesion, motility and vascular invasion with the purpose of identify a phenotypic profile able to predict HCC outcome. - Methods. Two tissue microarrays ( a western set composed of 98 HCV-correlated HCC cases and an eastern set composed of 136 HBV-correlated HCC cases) with clinicopathological information (aetiology, age, sex, grade, stage, micro and macro-vascular invasion, and patient’s follow up) was used to test the immunocytochemical expression of the following antigens: Ep-CAM, LAMA3, Osteopontin, PAK1, alpha- and beta-tubulin, CK19, GS, HSP70 e GPC3. - Results. Our data showed that tubulins are the only markers able to reveal prognostic information in both western and eastern population. Osteopontina, Pak1 and HSP70, singularly or associated with each other, are able to predict an unfavorable outcome in the eastern patients; the other markers should be validated with other studies. - Conclusions. Finally, 1) we observe different markers expression profile in the two different populations; 2) it may be a reflection of genetic, aetiology and epidemiology differences between the two populations; 3) grade and macroscopic vascular invasion are still the strongest pathological criteria on the multivariate analysis, so the new antibodies to be developed should be compared to these parameters.

The research project has been developed on the equine inflammatory respiratory diseases, which can be divided in Recurrent Airway Obstruction (RAO) and Inflammatory Airway Disease (IAD). The aim of this study was to investigate immune-related genes expression in the respiratory tract of IAD and RAO-affected horses. Clinical examination and endoscopy were performed. On the Broncho-Alveolar Lavage (BAL) fluid, obtained during endoscopy, cytological and microbiological analysis were performed to evaluate correlations between the gene expressions values and the clinical parameters. A first analysis was developed by real time RT-PCR comparing the gene expression profile of 10 immune-related target genes (IL-1ß, IL-6, IL-8, IL-13, IL-17, TNFa, INF?, TGF-ß1, NF?-ß and TRL 4) in the BAL of healthy horses and RAO-affected ones, on which sampling was performed twice within 15 days. The aim was to deepen the effects of the respiratory disease on the equine immune system and to assess a potential temporal evolution of the gene expression values and of the other parameters considered. In addition, biopsies of the bronchial tissue were obtained and subsequent, histological evaluation and gene expression analyses were performed. Six of the target genes showed a significant expression values increase in the RAO group compared to the control one. A positive statistical correlation between the amount of mucus in the airways and the expression of some genes investigated was found. Regarding inflammatory mediators expression in the biopsy tissue, neither of target genes was significantly differentially expressed between the RAO horses and the control group. The second part of the research project included also the study of IAD. On all horses, clinical investigations and assessments of gene expression profiles were carried out twice within 15 days, at the diagnosis moment and at the end of the pharmacological treatment. Considering the results of the first study, no biopsies of the respiratory tissue were performed. The development of a microarray platform specific for equine immune-related genes, provide a global view of the pathways involved in the IAD and RAO inflammatory response. The statistical analyses showed that 379 transcripts (55 up-regulated and 324 down-regulated) were significantly differentially expressed between the IAD group and control horses and 1763 genes (903 up-regulated and 860 down-regulated) between the RAO-affected horses and the healthy animals. Between IAD-affected horses and RAO animals, were showed significant differences of the respiratory rate at rest and of the amount of mucus in the airways. Some transcripts involved in the genesis, length and motility of the respiratory epithelium cilia, were down-regulated both in IAD and in RAO horses. In the IAD population, has been demonstrated the over-expression of genes coding for inflammatory mediators. Some of the transcripts up-regulated in the RAO group, are involved in the inflammatory response, bronchoconstriction, apoptosis and hypoxia pathway. In the same disease, some genes involved in the genesis of the protective muco-protein film of the respiratory epithelium were under-expressed. The analyses carried out by the software Gene Sets Enrichment Analysis (GSEA) showed that the pathway activated during human asthma, is also enriched in equine RAO, albeit marginally significant (False Discovery Rate <25%, p value 0.08 ). The low quality of the RNA extracted from the BAL of some samples, did not allow to reach a significant number of horses, considered before and after the pharmacological treatment, to assess the effect of the therapy on gene expression profiles. In conclusion, the present studies provided information about the immunological mechanisms activated during the most important equine respiratory diseases. In the future, the information obtained could lead to the development of new therapies for IAD and RAO, by the inhibition of molecules involved in the pathogenesis of these diseases, as is already done in human medicine. The involvement of the same pathway in human asthma and equine RAO, could suggest a possible role of horses as animal model for the study of human chronic respiratory diseases.
Il lavoro di ricerca svolto nell’arco dei tre anni di dottorato, è stato articolato in due progetti sviluppati nell’ambito delle malattie respiratorie su base infiammatoria che colpiscono gli equini. Tali patologie possono essere distinte in due grandi gruppi: Recurrent Airway Obstruction (RAO) ed Inflammatory Airway Disease (IAD). Lo scopo dei progetti di ricerca si è basato sull’indagine dei profili di espressione di geni immuno-correlati nell’albero respiratorio di cavalli affetti da IAD e RAO, in relazione ad un gruppo di controllo. Su tutti i soggetti, sono stati eseguiti gli esami clinici mirati alla valutazione dell’apparato respiratorio, l’esame endoscopico e l’esame citologico e microbiologico da Broncho-Alveolar Lavage (BAL), per valutare le potenziali correlazioni esistenti tra i profili di espressione genica ed i parametri clinici. Il primo progetto è stato sviluppato comparando cavalli sani con soggetti affetti da RAO, su cui i campionamenti sono stati ripetuti due volte nell’arco di 15 giorni, al fine valutare una potenziale evoluzione temporale dell’espressione genica e degli altri parametri considerati nella ricerca. Inoltre, sono state eseguite biopsie del tessuto bronchiale, sottoposto sia a valutazione istologica che ad analisi di espressione genica. Mediante real time RT-PCR, sono stati indagati i profili di espressione di 10 geni target immuno-correlati (IL-1ß, IL-6, IL-8, IL-13, IL-17, TNFa, INF?, TGF-ß1, NF?-ß e TRL 4), sei dei quali hanno dimostrato una aumento statisticamente significativo dei livelli di espressione nel gruppo RAO rispetto al gruppo di controllo. Le analisi statistiche condotte, hanno riscontrato una correlazione positiva tra la quantità di muco nelle vie aeree e l’ espressione di alcuni dei geni indagati. Non sono state evidenziate differenze di espressione, dei geni inclusi nello studio, tra i tessuti bioptici prelevati dai soggetti affetti da RAO e quelli ottenuti dal gruppo di controllo. Il secondo progetto di ricerca, è stato sviluppato ampliando la casistica dei cavalli affetti da RAO ed introducendo lo studio della IAD. Su tutti i soggetti, le indagini cliniche e le valutazioni dei profili di espressione genica sono state condotte sia al momento della diagnosi che al termine del trattamento farmacologico della durata di 15 giorni. Valutati i risultati del primo lavoro, non sono state eseguite biopsie del tessuto respiratorio. Lo sviluppo di una piattaforma microarray specifica per i geni immuno-correlati di cavallo ha permesso di ottenere una visione globale dei pathway coinvolti nella risposta infiammatoria delle due patologie. Le analisi statistiche effettuate hanno evidenziato una differenza di espressione significativa per 379 trascritti (di cui 55 sovra-espressi e 324 sotto-espressi) tra il gruppo IAD ed il gruppo di controllo e per 1763 geni (di cui 903 sovra-espressi e 860 sotto-espressi) tra i pazienti affetti da RAO ed i soggetti sani. Da un punto di vista clinico, sono state riscontrate differenze statisticamente significative sia della frequenza respiratoria a riposo che della quantità di muco presente nelle vie aeree dei cavalli affetti da IAD rispetto ai soggetti RAO. Tra i geni sotto-espressi nei due gruppi di cavalli affetti da malattia respiratoria, hanno acquistato importanza alcuni trascritti coinvolti nella genesi, lunghezza e motilità dell’apparato ciliare dell’epitelio respiratorio. Nella popolazione IAD, è stata dimostrata la sovra-espressione di geni codificanti per mediatori coinvolti nella risposta infiammatoria. I geni sovra-espressi nel gruppo RAO, caratterizzati da maggior rilievo, sono coinvolti nella risposta infiammatoria, nella broncocostrizione, nella via apoptotica e nel pathway dell’ipossia. Nella medesima patologia, si sono mostrati sotto-espressi anche alcuni geni coinvolti nella genesi del film muco-proteico di protezione dell’epitelio respiratorio. Lo studio condotto mediante Gene Set Enrichment Analysis (GSEA), ha evidenziato che il pathway attivato in corso di asma umano, viene arricchito anche nella patologia RAO equina, sebbene la significatività statistica sia marginale (False Discovery Rate < 25%, p value 0,08). Non è stato possibile valutare l’effetto della terapia farmacologica sui profili di espressione genica, poiché la bassa qualità dell’RNA estratto dal BAL di alcuni campioni non ha permesso di raggiungere un numero significativo di soggetti valutati prima e dopo il trattamento terapeutico. Gli studi effettuati hanno quindi permesso di far luce su alcuni dei meccanismi immunologici che stanno alla base delle patologie respiratorie equine di maggiore importanza veterinaria ed economica. In futuro, le informazioni ottenute, potrebbero condurre allo sviluppo di nuovi mezzi terapeutici per l’inibizione delle molecole coinvolte nello sviluppo di IAD e RAO, come già avviene in medicina umana. Infine, il coinvolgimento di un medesimo pathway nell’asma umano e nella RAO equina, potrebbe condurre all’utilizzo di tale specie come modello animale per lo studio delle patologie respiratorie croniche umane.

From the second half of past century to nowadays aquaculture keeps on being the fastest-growing animal-food-producing sector, so that it has provided the 46% of total food fish supply in 2010. However, we are faced with a few problems closely connected to some aspects that are still unknown in the biology of certain relevant species such as the gilthead seabream (Sparus aurata). Functional genomics can offer very well-grounded tools to get information about the molecular mechanisms which are involved in physiological processes whose consequences may be very high also from an economic point of view. The issue concerning how the marine organisms and populations react to climatic changes is a question of paramount importance which is still rather unsettled. The gilthead bream is very sensitive to low temperatures, so that it does not survive when temperature falls under 5°C. In fact, in winter time the breeding cause often huge economic losses to their owners since the mortality rate rises because of metabolic syndrome known as winter disease. In this study we considered the gene expression profiles of Sparus aurata individuals which have been exposed to low temperatures, in experimental conditions that could represent as realistic as possible the winter season. The gene expression profile can be used as a tool to link up the genotype to the physiology and to the phenotype. Moreover, the study looked into populations coming from regions with different climatic conditions (Veneto and Sicily), by assuming a different tolerance to cold exposition. Four groups of wild sea bream (120±16 g), coming in pairs from the two regions, were exposed for 21 days to two temperature treatments: 16 ± 0.3 °C (control groups) and 6.8 ± 0.3 °C (cold groups). Liver and gill samples were collected during acute (0, 6 and 24 hours) and chronic exposure (21 days). The gene expression profiles were analyzed using an oligo-nucleotide microarray technology, with about 19,715 ESTs. Results revealed a complex transcriptomic response to cold with many molecular pathways involved among which: lipid and carbohydrate metabolism, regulation of heat shock proteins (HSPs) and other protein chaperones, protein degradation and repair, regulation of cell death, RNA and DNA metabolism, immune response. The earliest transcriptional response is linked to oxidative stress and anti-oxidant/survival cell response, suggesting an immediate disturbance of systemic oxygen balance. The largest transcriptional difference between cold and control groups occurred during long-term exposure, involving primarily several genes of lipid metabolism with a role in the re-allocation of energy sources and immune-related genes indicating an immunosuppressive effect of cold exposure. The data on the liver and gill transcriptome of the gilthead sea bream exposed to cold provide a starting point to investigate physiological mechanisms underlying long term cold adaptation in fish and to address future research for the identification of cold tolerant S. aurata strain for aquaculture.
Dalla seconda metà del secolo scorso ad oggi l‟acquacoltura continua ad essere il settore delle produzioni animali in più rapida crescita, con il 46% di pesce fornito sul totale consumato nel 2010. Rimangono, tuttavia, problematiche strettamente legate ad aspetti ancora sconosciuti nell‟ambito della biologia di alcune specie d‟interesse come l‟orata comune (Sparus aurata). La genomica funzionale può fornire validi strumenti per ottenere informazioni sui meccanismi molecolari coinvolti nei processi fisiologici importanti anche da un punto di vista economico. Come le popolazioni e le specie marine reagiscono ai cambiamenti climatici è una questione di importanza centrale ancora non del tutto risolta. L‟orata comune risente fortemente del freddo, non sopravvivendo a temperature inferiori ai 5°C e spesso, durante l‟inverno, gli allevamenti subiscono ingenti danni economici per l‟elevata mortalità data dalla sindrome metabolica winter disease. In questo studio sono stati valutati i profili di espressione genica di individui di S. aurata esposti alle basse temperature, in condizioni sperimentali che fossero il più realistiche possibile con la stagione invernale. Il profilo di espressione genica può servire come strumento per legare il genotipo alla fisiologia e al fenotipo. Sono state, inoltre, esaminate popolazioni provenienti da regioni con condizioni climatiche diverse, Veneto e Sicilia, ipotizzando una differente tolleranza al freddo. Quattro gruppi di orate (120±16 g), provenienti a coppie dalle due regioni, sono state esposte per 21 giorni a due trattamenti di temperatura: 16 ± 0,3 °C (gruppi di controllo) e 6,8 ± 0,3 °C (gruppi dei trattati). Campioni di fegato e branchia sono stati raccolti durante esposizione acuta (0, 6 e 24 ore) e cronica (21 giorni). I profili di espressione sono stati analizzati usando un microarray a oligo-nucleotidi con circa 19.715 geni. I risultati hanno rivelato una risposta trascrizionale complessa per la risposta al freddo, con numerosi pathway coinvolti: metabolismo di lipidi e carboidrati, heat shock protein (HSP) e chaperoni, degradazione proteica, apoptosi, metabolismo di RNA e DNA, risposta immunitaria. La prima risposta è legata allo stress ossidativo, suggerendo un disturbo immediato del bilancio dell‟ossigeno a livello sistemico, mentre le più grandi differenze trascrizionali tra trattati e controlli si rilevano durante l‟esposizione a lungo termine, e coinvolgono principalmente geni del metabolismo lipidico per la ridistribuzione delle riserve energetiche e geni dell‟immunità per l‟importante effetto immuno-soppressivo del freddo. I dati del trascrittoma di branchia e fegato di orate esposte alle basse temperature forniscono un punto di partenza per indagare i meccanismi fisiologici sottostanti l‟adattamento al freddo a lungo termine nei pesci e per indirizzare ricerche future volte all‟identificazione di ceppi di S. aurata resistenti al freddo in acquacoltura.

Expression profiles are increasingly used in cancer research, and a growing number of microarray data is available in public databases such as Gene Expression Omnibus (GEO) and ArrayExpress. Therefore, it is possible to compare studies with similar research aims through a meta-analysis approach. Colorectal cancer (CRC) is a very common disease and represents a fundamental biological model of tumorigenesis. Many experiments comparing the expression profiles of normal colonic mucosa and colorectal cancer (CRC) samples have been conducted and results are available. Using A-MADMAN, a web application that allows the retrieval, annotation, organization and analysis of public available gene expression datasets, we downloaded from the GEO and collected the largest-collected normal colon, adenoma and primary colorectal tumor, with and without metastasis, gene expression dataset which includes 1,795 experiments pertaining to 27 distinct series, obtained from two different generations of Affymetrix, HG-U133A and HG-U133 plus2.0. After we performed the quality control of the downloaded data and have designed three different workflows for signal reconstruction, we conducted preliminary analysis of differential expression. In this first project we focused on right and left colon differences. In particularly, we analyzed normal, adenoma and tumor expression profiles focusing on changes among topologically different samples, under the hypothesis that the use of site-specific normal tissue as reference for the study of right and left tumor samples may facilitate to shed light on specific cell functions, pathways and regulatory circuits altered in each tumor type, and help deepening our understanding of this complex disease. From preliminary data, we concluded that the workflow based on the virtual chips production and on a chip-specific background correction is the most robust. Higher numbers of differentially expressed genes were identified when comparing, respectively, tumor and adenoma versus normal tissue (647 and 683 genes). These tissue samples were then pairwise compared by topology to identify those genes that are differentially expressed, when the matched by topology set of normal samples is considered. We found 72 genes colon left-specific and 1183 colon-right specific in the comparison from adenoma with normal tissues, and 46 genes and 69 genes, respectively, in colon left- and colon right- specific in the comparison from CRC and adenoma tissues. We focused also to the functional information associated to gene expression differences between tumor and normal tissue samples, compared considering only right or left samples, to identify specific functions and pathways involved with cancer present, such as DNA repair, adhesion and cell interaction. In the second project we reconstructed regulatory networks of colon cancer by integrating gene expression, alternative splicing and microRNA (miRNAs) expression data. We carried out a genome-wide integrative analysis of miRNA and genes and exon expression profiles in tissues of 55 colon cancer patients, comprehensive of normal colon mucosa, primary tumor and liver metastasis. GeneChip Human Exon 1.0 ST (Affymetrix) and Affymetrix GeneChip miRNA arrays have been used to obtain high quality gene, exon and miRNA expression quantification respectively. We analyzed both differential gene expression and alternative splicing applying AltAnalyze software to exon-level analysis, identifying 33,740 genes involved in alternative splicing events. We considered a custom set of expression signals of 449,810 probesets, using MIDAS and FIRMA statistics, and we combined the results with MMBGX software output, to identify a more restricted, but perhaps more robust set of candidate alternative splicing events, possibly relevant for colon cancer biology. When comparing liver metastasis with normal colon tissue whereas 182 genes were identified; comparable numbers of genes were identified when comparing metastasis versus colorectal tumor and colorectal tumor versus normal tissue (51 and 10 respectively). We validated two genes with alternative splicing events, VCL and CALD1. We then identified respectively 62, 63 and 11 differentially expressed miRNAs in tumors and metastases compared to normal tissue, and comparing metastasis with primary tumors. To assess the reproducibility and robustness of the miRNA signature identified, we measured by qRT-PCR the expression of 5 miRNA in all samples (hsa miR-150, hsa miR-10b, has miR-146a, miR-210 and has miR-122). For each considered contrast, KEGG pathways enriched in genes resulting supported target of differentially expressed miRNAs were identified. The integrated analysis of miRNA and genes expression profiles with target predictions was performed with the purpose of reconstruct post-transcriptional regulatory networks governing tumor and metastases development. Considering expression profiles in the same set of samples of 305 miRNAs and 12,748 genes with variable expression profile, we reconstructed post-transcriptional regulatory networks involving modulated miRNAs. In particular, considering the network associated to the tumor versus normal contrast, we experimentally validated miR-145 – c-Myc and miR-182 – ENTPD5 relationships. This latter is new and may have a relevant pathogenetic role. Looking at our results, we can say that the regulatory signatures that affect tumor progression are complex and difficult to interpret. They involve interactions involving different modulated miRNAs that control the expression of multiple genes belonging to the same pathway in different ways, and alternative transcripts of genes that are differentially expressed in normal tissues, tumors and metastases.
Nella ricerca oncologica sono sempre più utilizzati i profili di espressione genica e un numero crescente di dati di microarray è disponibile in database pubblici come NCBI GEO e ArrayExpress. Diventa quindi possibile il confronto di studi con obiettivi di ricerca simili attraverso approcci di “meta-analisi”. Il cancro colorettale (CRC) rappresenta un fondamentale modello biologico di tumorigenesi, oltre ad essere una patologia molto diffusa. Sono a disposizione nei database pubblici molti esperimenti sul CRC, che confrontano mucosa normale con mucosa tumorale del colon attraverso metodologia microarray. Noi abbiamo utilizzato A-MADMAN, un’applicazione web open source di supporto per la meta-analisi di dati grezzi d’espressione genica ottenuti con microarray, per scaricare dal database dell’NCBI Gene Expression Omnibus (GEO), 27 collezioni di esperimenti per un totale di 1045 campioni ottenuti da mucosa normale, adenoma e CRC con e senza metastasi di colon-retto su cui erano state eseguite analisi di espressione genica con tecnologia gene chip Affymetrix. Dopo aver effettuato un controllo di qualità dei dati scaricati e aver disegnato 3 diversi flussi di lavoro per la ricostruzione del segnale d’espressione, sono state condotte delle analisi preliminari di espressione differenziale. In questo primo progetto ci siamo focalizzati sulle differenze molecolari sito–specifiche. Abbiamo quindi analizzato campioni di mucosa sana, di adenoma e di CRC suddivisi per localizzazione in colon destro e colon sinistro. Prendendo come riferimento il tessuto normale destro e sinistro, abbiamo ipotizzato che è possibile discriminare pattern di espressione genica tumorale sede specifica. Dal disegno dei tre flussi di lavoro abbiamo dedotto che il flusso di lavoro basato sulla generazione di chip virtuali e su una correzione del background chip-specifica è il più affidabile. Abbiamo identificato 647 geni differenzialmente espressi confrontando tessuto tumorale con tessuto normale e 683 geni nel confronto adenoma con tessuto normale, poi abbiamo individuato 72 geni nel confronto adenoma-tessuto normale specifici del colon sinistro e 1183 geni nello stesso confronto, colon destro specifici, mentre per quanto riguarda il confronto CRC e adenoma, 46 geni sono colon sinistro specifici e 69 sono colon destro specifici. Abbiamo inoltre trovato termini funzionali e pathway diversi sovra-rappresentati in colon destro rispetto a colon sinistro e rispetto ai corrispondenti normali. Nel secondo progetto, abbiamo ricostruito una rete regolatoria riguardante il CRC e la metastasi epatica da CRC integrando dati di originali di espressione genica, di splicing alternativo e di espressione di microRNA (miRNA). Lo studio si basa sulla raccolta di biopsie di tumore primario al colon, mucosa adiacente normale e metastasi al fegato di 55 pazienti, che sono state analizzate utilizzando piattaforme Affymetrix per l'analisi di espressione genica/esonica (GeneChip Human Exon 1,0 ST), e di microRNA (GeneChip® miRNA Array ). Analizzando l’ espressione differenziale a livello genico ed esonico mediante il software AltAnalyze identificando 33.740 geni coinvolti in probabili eventi di splicing alternativo. Integrando i dati risultati dalle statistiche di questo programma con i risultati ottenuti con un software basato su un approccio Bayesiano, MMBGX, abbiamo identificato una lista più ristretta ma anche più robusta di candidati eventi di splicing possibilmente rilevanti per la formazione e la progressione del CRC. Confrontando metastasi epatiche con tessuto normale del colon sono stati identificati 182 geni, mentre un numero inferiore di geni sono stati identificati nel confronto contro le metastasi del tumore colorettale e del tumore del colon-retto rispetto al tessuto normale (51 e 10 rispettivamente). A partire da questi risultati, abbiamo scoperto il coinvolgimento di trascritti alternativi di due geni, VCL e CALD1, nella progressione tumorale. In parallelo, sono stati identificati microRNA modulati in seguito allo sviluppo del tumore e della metastasi, trovandone rispettivamente 62, 63 e 11 differenzialmente espressi nel tumore rispetto al normale, nella metastasi rispetto al normale e nella metastasi rispetto al tumore. Abbiamo quindi confermato la robustezza dei risultati validando cinque miRNA presenti nella lista dei differenzialmente espressi (hsa miR-150, hsa miR-10b, has miR-146a, miR-210 and has miR-122) mediante RT-PCR. Per ogni contrasto considerato, sono stati identificati i KEGG pathway modulate e quindi sotto putativamente controllate dai miRNA. Grazie all’analisi integrata di profili di espressione di miRNA e dei loro geni target anti-correlati, sono state definite le principali reti di regolazione post-trascrizionale coinvolte nella cancerogenesi. Particolarmente rilevante è la rete che coinvolge il sottoinsieme dei miRNA differenzialmente espressi nel tumore rispetto al normale. Tra le interazioni inferite, abbiamo convalidato sperimentalmente le relazioni miR-145 - c-Myc e miR-182 - ENTPD5. Quest’ultima rappresenta una relazione nuova, il cui ruolo patogenetico può essere rilevante. Dai nostri risultati possiamo concludere che le vie di regolazione che interessano la progressione tumorale sono complesse e difficili da interpretare, che implicano interazioni che coinvolgono miRNA diversamente modulati che agiscono in diversi modi sull’espressione di più geni appartenenti ad una stessa pathway e da trascritti alternativi di geni che vengono espressi in modo differenziale nei tessuti sani, nei tumori e nelle metastasi.

Cluster headache (CH) and bipolar disorder (BD) are pathological conditions affecting 1% and 1.5% of the general population, respectively. Albeit the pathogenesis has not yet been completely elucidated, family and twin studies have suggested a genetic basis for both disorders, with an estimated heritability of 80% for BD and up to 60% for CH. Though BD and CH are very different diseases, they show important clinical similarities, such as a temporal pattern of disturbances, dysregulation of the wake-­‐sleep cycle, neuroendocrine derangements, and more important positive clinical response to lithium and valproate treatments in a significant proportion of treated patients. In the present study, we sought to explore whether BD and CH patients responders to lithium share common molecular pathways potentially involved in predisposing to positively respond to prophylactic lithium treatment. To this aim, we carried out a transcriptome study in lymphoblastoid cell lines from 10 BD type I patients, responders to lithium, 8 CH patients responders to lithium treatment and 10 healthy subjects (CO). Expression profiles were measured by Affymetrix GeneChip Human Gene ST 1.0 covering 36,079 transcripts. Expression levels of BD and CH patients were compared with CO using a t-­‐test, in order to identify commonly dysregulated genes. Pathway analysis was performed based on the hypergeometric test for over representation of specific Kyoto Encyclopedia of Genes and Genomes (KEGG). A total of 544 and 1172 genes were differentially expressed in BD versus CO and CH versus CO respectively. Filtering criteria were based on corrected p value < 0.05 and a Fold Change (FC) ≥ |1.3|. Among these genes, 314 were commonly altered both in CH and BD compared to CO. The most significant dysregulated gene in BD and CH was RNA binding motif (RNP1, RRM) protein 3 (RBM3), a gene implicated in sleep regulation and in the temperature entrained circadian gene expression (corrected p value of 6,30x 10-­‐09 in BD vs CO and 1,88x 10-­‐09 in CH vs CO). Pathway analysis showed that Protein processing in endoplasmic reticulum pathway was one of the most significantly enriched in BD and CH when compared to CO. In conclusion, data from this pilot microarray study may provide useful and relevant information for a better understanding of the molecular underpinnings of lithium response and on the neurobiology of BD and CH.

Thesis (Ph. D.)--Ohio State University, 2005.
Title from first page of PDF file. Document formatted into pages; contains ix, 146 p.; also includes graphics (some col.) Includes bibliographical references (p. 145-146). Available online via OhioLINK's ETD Center

DNA microarrays are a relatively new technology for assessing the expression levels of thousands of genes simultaneously. Researchers hope to find genes that are differentially expressed by hybridizing cDNA from known treatment sources with various genes spotted on the microarrays. The large number of tests involved in analyzing microarrays has raised new questions in multiple testing. Several approaches for identifying differentially expressed genes have been proposed. This paper considers two: (1) a mixed models approach, and (2) the Signiffcance Analysis of Microarrays.

In microarray analysis, people are interested in those features that have different characters in diseased samples compared to normal samples. The usual p-value method of selecting significant genes either gives too many false positives or cannot detect all the significant features. The False Discovery Rate (FDR) method controls false positives and at the same time selects significant features. We introduced Benjamini's method and Storey's method to control FDR, applied the two methods to human Meningioma data. We found that Benjamini's method is more conservative and that, after the number of the tests exceeds a threshold, increase in number of tests will lead to decrease in number of significant genes. In the second chapter, we investigate ways to search interesting gene expressions that cannot be detected by linear models as t-test or ANOVA. We propose a novel approach to use quadratic logistic regression to detect genes in Meningioma data that have non-linear relationship within phenotypes. By using quadratic logistic regression, we can find genes whose expression correlates to their phenotypes both linearly and quadratically. Whether these genes have clinical significant is a very interesting question, since these genes most likely be neglected by traditional linear approach.

Normal kidney and urinary tract development is a complex process, regulated by a strict space-time-corrected sequential activation of a cascade of genes encoding transcription factors, growth factors, cell death/proliferation factors and adhesion molecules. An alteration disrupting this sequential gene expression may cause a defective ureteric bud-to-metanephric mesenchyme cross-talk that results in a renal and urinary tract developmental abnormality (congenital anomalies of kidney and urinary tract - CAKUT). Phenotype severity depends on the stage of nephrogenesis in which the alteration of the developmental program occurs, thus renal dysplasia is the most severe manifestation. However, little is known about CAKUT pathogenesis. The recent advent of microarray technology provided an unique tool to identify genes potentially involved in the pathogenesis of several diseases. During the first stage of this research, we applied the microarray technique to study gene expression profiles of primary renal cell cultures, using an array composed by 21329 oligonucleotides. The aim was to identify potential biomarkers of renal dysplasia. Four genes seemed to be more interesting (UPK1B, SOX11, SPRY1, MMP2). We analysed the expression of these four genes using Real Time PCR on RNA extracted from renal tissue samples of 10 patients with a histological picture of renal dysplasia and 10 with histologically normal renal tissue. Mutation analysis of SPRY1 gene, whose murine homologue is hugely involved in the regulation of GDNF growth factor's expression during ureteric branching, was carried out on 27 patients with renal duplicity. Mutation analysis identified 2 new genomic variants - whose frequency was analysed in a control population - that may be "genomic variants involved in splicing" (SpaGVs). Our research results allow to hypothesize that SPRY1 gene may be involved in the pathogenesis of kidney and urinary tract developmental diseases.

In recent years, the development of new genomic technologies has allowed for the investigation of many regulatory epigenetic marks besides expression levels, on a genome-wide scale. As the price for these technologies continues to decrease, study sizes will not only increase, but several different assays are beginning to be used for the same samples. It is therefore desirable to develop statistical methods to integrate multiple data types that can handle the increased computational burden of incorporating large data sets. Furthermore, it is important to develop sound quality control and normalization methods as technical errors can compound when integrating multiple genomic assays. DNA methylation is a commonly studied epigenetic mark, and the Infinium HumanMethylation450 BeadChip has become a popular microarray that provides genome-wide coverage and is affordable enough to scale to larger study sizes. It employs a complex array design that has complicated efforts to develop normalization methods. We propose a novel normalization method that uses a set of stable methylation sites from housekeeping genes as empirical controls to fit a local regression hypersurface to signal intensities. We demonstrate that our method performs favorably compared to other popular methods for the array. We also discuss an approach to estimating cell-type admixtures, which is a frequent biological confound in these studies. For data integration we propose a gene-centric procedure that uses canonical correlation and subsequent permutation testing to examine correlation or other measures of association and co-localization of epigenetic marks on the genome. Specifically, a likelihood ratio test for general association between data modalities is performed after an initial dimension reduction step. Canonical scores are then regressed against covariates of interest using linear mixed effects models. Lastly, permutation testing is performed on weighted correlation matrices to test for co-localization of relationships to physical locations in the genome. We demonstrate these methods on a set of developmental brain samples from the BrainSpan consortium and find substantial relationships between DNA methylation, gene expression, and alternative promoter usage primarily in genes related to axon guidance. We perform a second integrative analysis on another set of brain samples from the Stanley Medical Research Institute.

Tissue microarrays (TMAs) are a high-throughput technique that facilitates the survey of very large numbers of tumours, important both in clinical and research applications. However, the assessment of stained TMA sections is laborious and still needs to be carried manually, constituting a bottleneck in the pathologist?s work-flow. This process is also prone to perceptual errors and observer variability.Thus, there is strong motivation for the development of automated quantitative analysis of TMA image data. The analysis of breast TMA sections subjected to nuclear immunostaining begins with the classification of each spot as to the maintype of tissue that it contains, namely tumour, normal, stroma, or fat. Tumour and normal spots are then assigned a so-called quickscore composed of a pair or integer values, the first reflecting the proportion of epithelial nuclei that are stained, and the second reflecting the strength of staining of those nuclei. In this work, an approach was developed to analyse breast TMA spots subjectedto progesterone receptor immunohistochemistry. Spots were classified into their four main types through a method that combined a bag of features approachand classifiers based on either multi-layer perceptrons or latent Dirichlet allocation models. A classification accuracy of 74.6 % was achieved. Tumour and normal spots were scored via an approach that involved the computation of global features formalising the quickscore values used by pathologists, and the use of Gaussian processes for ordinal regression to predict actual quickscores based on global features. Mean absolute errors of 0.888 and 0.779 were achieved in the prediction of the first and second quickscore values, respectively. By setting thresholds on prediction confidence, it was possible to classify and score fractions of spots with substantially higher accuracies and lower mean absolute errors. Amethod for the segmentation of TMA spots into regions of different types was also investigated, to explore the generative nature of latent Dirichlet allocation models.

Neste trabalho é proposto um estudo comparativo de alguns métodos de Agrupamento (Hierárquico, K-médias e Self-Organizing Maps) e de Classificação (K-Vizinhos, Fisher, Máxima Verossimilhança, Aggregating e Regressão Local), os quais são apresentados teoricamente. Tais métodos são testados e comparados em conjuntos de dados reais, gerados com a técnica de Microarray. Esta técnica permite mensurar os níveis de expressão de milhares de genes simultaneamente, possibilitando comparações entre amostras de tecidos pelos perfis de expressão. É apresentada uma revisão de conceitos básicos relacionados ao processo de normalização, sendo este uma das primeiras etapas da análise deste tipo de conjunto de dados. Em particular, estivemos interessados em encontrar pequenos grupos de genes que fossem ?suficientes? para distinguir amostras em condições¸ biológicas diferentes. Por fim, é proposto um método de busca que, dado os resultados de um experimento envolvendo um grande número de genes, encontra de uma forma eficiente os melhores classificadores.
In this work we propose a comparative study of some clustering methods (Hierarchic, K -Means and Self-Organizing Maps) and some classification methods (K-Neighbours, Fisher, Maximum Likelihood, Aggregating and Local Regression), which are presented teoretically. The methods are tested and compared based on the analysis of some real data sets, generated from Microarray experiments. This technique allows for the measurement of expression levels from thousands of genes simultaneously, thus allowing the comparative analysis of sample of tissues in relation to their expression profile. We present a review of basic concepts regarding normalization of microarray data, one of the first steps in microarray analysis. In particular, we were interested in finding small groups of genes that were ?sufficient? to identify samples originating from different biological conditions. Finally, a search method is proposed, which will find efficiently the best classifiers from the results of an experiment involving a huge number of genes.

This work concerns with data analysis of DNA microarrays by using cluster analysis. It explains biological terms - gene expression and DNA microarray. Next, it contains mathematical and informatical description of clustering methods and describes a way to apply these methods to microarrays data. Next, the work contains implementation's detail of clustering methods k-means, DBSCAN and introduces an original clustering algorithm Strom++. Then, description of implementation and application manual follow. Finally, accomplished results are evaluated.

Nell’ultimo decennio la scoperta dei microRNA ha messo in luce un nuovo e fine meccanismo di regolazione post-trascrizionale, che interviene in molti processi quali lo sviluppo, il differenziamento la proliferazione e la morte cellulare. Inoltre, numerose evidenze hanno dimostrato il coinvolgimento dei microRNA in diverse patologie. Nella maggior parte degli studi effettuati, e' stata utilizzata la tecnologia microarray per identificare i microRNA coinvolti nei meccanismi patogenetici, ma anche per ottenere dei profili di espressione caratteristici di patologia con valore diagnostico o prognostico. Questi studi suggeriscono che l'analisi dei profili di espressione dei microRNA può essere considerata uno strumento utile per comprendere quale ruolo essi svolgono nella regolazione dei processi fisiopatologici. In questo lavoro abbiamo studiato il profilo di espressione di microRNA, mediante tecnologia microarray in due diversi modelli sperimentali: 1) un modello in vitro, utile per la comprensione dei meccanismi molecolari alla base della risposta immunitaria; 2) un modello in vivo, idoneo per lo studio della patogenesi di una patologia epatica molto diffusa nota come NAFLD ("Non-alcoholic fatty liver disease"). E' di recente scoperta la relazione tra infiammazione, immunità innata e microRNA, che sono descritti essere coinvolti nella regolazione della risposta cellulare all’infezione microbica. Noi abbiamo, così, analizzato lo specifico profilo di espressione di microRNA in cellule dendritiche umane, utilizzando un modello in vitro di stimolazione e attivazione mediante agonisti di differenti recettori Toll-like (TLRs): R848/Resiquimod, ligando del TLR7/8; LPS, ligando del TLR4; e poly(I:C), ligando del TLR3. Questa analisi ha permesso di identificare gruppi di microRNA espressi specificamente in risposta a determinati stimoli e puo' risultare utile per chiarire il possibile ruolo dei microRNA nei meccanismi attraverso i quali le cellule dendritiche discriminano i diversi patogeni. La steatosi epatica di origine non-alcolica, o NAFLD, è una patologia emergente caratterizzata da un ampio spettro di condizioni epatiche: dalla semplice steatosi, alla steatoepatite con fibrosi più o meno avanzata (NASH, non-alcoholic steatohepatitis). La patologia nelle sue forme più gravi può evolvere fino alla cirrosi e all’epatocarcinoma. La NAFLD ha una complessa eziopatogenesi ancora poco chiara. Per individuare i possibili meccanismi molecolari coinvolti nello sviluppo della NAFLD abbiamo utilizzato un modello animale, capace di riprodurre i vari aspetti della patologia umana. In prticolare, in questo studio abbiamo effettuato l’analisi dei profili di espressione dei microRNA nel tessuto epatico di ratti sottoposti a diversi regimi dietetici. I nostri risultati hanno dimostrato che il trattamento con i diversi regimi ipercalorici causa un aumento significativo del peso corporeo e del fegato, e di alcuni parametri metabolici rispetto agli animali controllo, come anche differenti danni epatici. L’analisi dei microRNA ha dimostrato la significativa deregolazione di 3 microRNA down-regolati (miR-122, miR-451 e miR-27a) e 3 up-regolati (miR-200a, miR-200b e miR-429) negli animali sottoposti alle diete ipercaloriche rispetto alla dieta standard. Fra i potenziali bersagli di tali microRNA emergono alcune molecole coinvolte nel controllo dell’apoptosi e dell'infiammazione, ma soprattutto proteine del segnale intracellulare, e del metabolismo lipidico e glucidico.
Over the last decade, the discovery of microRNAs revealed a new mechanism of post-transcriptional regulation. MicroRNAs are involved in many biological processes such as development, differentiation, proliferation and cell death. Moreover, several evidences showed the pathogenic role of microRNAs in various diseases. A lot of studies used microarray technology to identify miRNAs involved in the pathogenesis, but also to obtain the expression pattern characteristic of pathology with diagnostic or prognostic assessment. These studies suggest that profiling of microRNAs may be used to understand the role they play in regulating pathophysiological processes. In this work we employed microarray technology to investigate the expression profile of microRNAs in two different experimental models: 1) an in vitro model, useful for understanding the molecular mechanisms underlying the immune response, 2) a in vivo model, suitable for studying the pathogenesis of non-alcoholic fatty liver disease, also known as NAFLD. Recently, has been explored the relationship between inflammation, innate immunity and microRNAs, which are described to be involved in regulating cellular response to microbial infection. Thus, we identified the specific expression profile of microRNAs in human dendritic cells, using an in vitro model of stimulation and activation by agonists of different Toll-like Receptors (TLRs): R848/Resiquimod, ligand of TLR7/8; LPS, ligand of TLR4; and poly(I: C), ligand of TLR3. Analysis of expression profiles identified groups of miRNAs expressed specifically in response to treatments with LPS, R848, or their combination with respect to control dendritic cells. This analysis will help to clarify their possible role in mechanisms of dendritic cells to discriminate pathogens. The non-alcoholic fatty liver disease or NAFLD is an emerging disease characterized by a wide spectrum of liver conditions from simple steatosis, steatohepatitis with or without fibrosis (NASH, non-alcoholic steatohepatitis). The etiopathogenesis of NAFLD is complex and still unclear. To identify possible molecular mechanisms involved in the development of NAFLD, we used an animal model, able to reproduce various aspects of human pathology. In this study we performed the analysis of microRNAs expression profiles in liver tissue of rats subjected to different diets. Our results showed that treatment with different ipercaloric regimens caused a significant increase in body weight and liver, and some metabolic parameters, compared to control animals, as well as different liver damage. The analysis of microRNAs showed the significant downregulation of three microRNAs (miR-122, miR-451 and miR-27a) and the up-regulation of other three microRNAs (miR-200A, miR-429 and miR-200B) in animals treated with ipercaloric diets respect to those with a standard diet. Among the potential targets of these microRNAs we found some molecules involved in the regulation of apoptosis and inflammation, but also intracellular signaling proteins, and lipid and glucose metabolism.

Microarray technology enables researchers to measure the expression levels of thousands of genes simultaneously to understand relationships between genes, extract pathways, and in general understand a diverse amount of biological processes such as diseases and cell cycles. While microarrays provide the great opportunity of revealing information about biological processes, it is a challenging task to mine the huge amount of information contained in the microarray datasets. Generally, since an accurate model for the data is missing, first a clustering algorithm is applied and then the resulting clusters are examined manually to find genes that are related with the biological process under inspection. We need automated methods for this analysis which can be used to eliminate unrelated genes from data and mine for biologically important genes. Here, we introduce a general methodology which makes use of traditional clustering algorithms and involves integration of the two main sources of biological information, Gene Ontology and interaction networks, with microarray data for eliminating unrelated information and find a clustering result containing only genes related with a given biological process. We applied our methodology successfully on a number of different cases and on different organisms. We assessed the results with Gene Set Enrichment Analysis method and showed that our final clusters are highly enriched. We also analyzed the results manually and found that most of the genes that are in the final clusters are actually related with the biological process under inspection.

Single nucleotide polymorphisms (SNPs) are DNA sequence variations, occurring when a single nucleotide –A, T, C or G – is altered. Arguably, SNPs account for more than 90% of human genetic variation. Dr. Tebbutt's laboratory has developed a highly redundant SNP genotyping assay consisting of multiple probes with signals from multiple channels for a single SNP, based on arrayed primer extension (APEX). The strength of this platform is its unique redundancy having multiple probes for a single SNP. Using this microarray platform, we have developed fully-automated genotype calling algorithms based on linear models for individual probe signals and using dynamic variable selection at the prediction level. The algorithms combine separate analyses based on the multiple probe sets to give a final confidence score for each candidate genotypes. Our proposed classification model achieved an accuracy level of >99.4% with 100% call rate for the SNP genotype data which is comparable with existing genotyping technologies. We discussed the appropriateness of the proposed model related to other existing high-throughput genotype calling algorithms. In this thesis we have explored three new ideas for classification with high dimensional data: (1) ensembles of various sets of predictors with built-in dynamic property; (2) robust classification at the prediction level; and (3) a proper confidence measure for dealing with failed predictor(s). We found that a mixture model for classification provides robustness against outlying values of the explanatory variables. Furthermore, the algorithm chooses among different sets of explanatory variables in a dynamic way, prediction by prediction. We analyzed several data sets, including real and simulated samples to illustrate these features. Our model-based genotype calling algorithm captures the redundancy in the system considering all the underlying probe features of a particular SNP, automatically down-weighting any ‘bad data’ corresponding to image artifacts on the microarray slide or failure of a specific chemistry. Though motivated by this genotyping application, the proposed methodology would apply to other classification problems where the explanatory variables fall naturally into groups or outliers in the explanatory variables require variable selection at the prediction stage for robustness.

The Top Scoring Pair (TSP) classifier, based on the notion of relative ranking reversals in the expressions of two marker genes, has been proposed as a simple, accurate, and easily interpretable decision rule for classification and class prediction of gene expression profiles. We introduce the AUC-based TSP classifier, which is based on the Area Under the ROC (Receiver Operating Characteristic) Curve. The AUCTSP classifier works according to the same principle as TSP but differs from the latter in that the probabilities that determine the top scoring pair are computed based on the relative rankings of the two marker genes across all subjects as opposed to for each individual subject. Although the classification is still done on an individual subject basis, the generalization that the AUC-based probabilities provide during training yield an overall better and more stable classifier. Through extensive simulation results and case studies involving classification in ovarian, leukemia, colon, and breast and prostate cancers and diffuse large b-cell lymphoma, we show the superiority of the proposed approach in terms of improving classification accuracy, avoiding overfitting and being less prone to selecting non-informative pivot genes. The proposed AUCTSP is a simple yet reliable and robust rank-based classifier for gene expression classification. While the AUCTSP works by the same principle as TSP, its ability to determine the top scoring gene pair based on the relative rankings of two marker genes across {\em all} subjects as opposed to each individual subject results in significant performance gains in classification accuracy. In addition, the proposed method tends to avoid selection of non-informative (pivot) genes as members of the top-scoring pair.\\ We have also proposed the use of the AUC test statistic in order to reduce the computational cost of the TSP in selecting the most informative pair of genes for diagnosing a specific disease. We have proven the efficacy of our proposed method through case studies in ovarian, colon, leukemia, breast and prostate cancers and diffuse large b-cell lymphoma in selecting informative genes. We have compared the selected pairs, computational cost and running time and classification performance of a subset of differentially expressed genes selected based on the AUC probability with the original TSP in the aforementioned datasets. The reduce sized TSP has proven to dramatically reduce the computational cost and time complexity of selecting the top scoring pair of genes in comparison to the original TSP in all of the case studies without degrading the performance of the classifier. Using the AUC probability, we were able to reduce the computational cost and CPU running time of the TSP by 79\% and 84\% respectively on average in the tested case studies. In addition, the use of the AUC probability prior to applying the TSP tends to avoid the selection of genes that are not expressed (``pivot'' genes) due to the imposed condition. We have demonstrated through LOOCV and 5-fold cross validation that the reduce sized TSP and TSP have shown to perform approximately the same in terms of classification accuracy for smaller threshold values. In conclusion, we suggest the use of the AUC test statistic in reducing the size of the dataset for the extensions of the TSP method, e.g. the k-TSP and TST, in order to make these methods feasible and cost effective.

Recent developments in automation and novel experimental techniques have led to the accumulation of vast amounts of biological data and the emergence of numerous databases to store the wealth of information. Consequentially, bioinformatics have drawn considerable attention, accompanied by the development of a plethora of tools for the analysis of biological data. DNA microarrays constitute a prominent example of a high-throughput experimental technique that has required substantial contribution of bioinformatics tools. Following its popularity there is an on-going effort to integrate gene expression with other types of data in a common analytical approach. Pathway based microarray analysis seeks to facilitate microarray data in conjunction with biochemical pathway data and look for a coordinated change in the expression of genes constituting a pathway. However, it has been observed that genes in a pathway may show variable expression, with some appearing activated while others repressed. This thesis aims to add some contribution to pathway based microarray analysis and assist the interpretation of such observations, based on the fact that in all organisms a substantial number of genes take part in more than one biochemical pathway. It explores the hypothesis that the expression of such genes represents a net effect of their contribution to all their constituent pathways, applying statistical and data mining approaches. A heuristic search methodology is proposed to manipulate the pathway contribution of genes to follow underlying trends and interpret microarray results centred on pathway behaviour. The methodology is further refined to account for distinct genes encoding enzymes that catalyse the same reaction, and applied to modules, shorter chains of reactions forming sub-networks within pathways. Results based on various datasets are discussed, showing that the methodology is promising and may assist a biologist to decipher the biochemical state of an organism, in experiments where pathways exhibit variable expression.

Selection of signi?cant genes via expression patterns is important in a microarray problem. Owing to small sample size and large number of variables (genes), the selection process can be unstable. This research proposes a hierarchical Bayesian model for gene (variable) selection. We employ latent variables in a regression setting and use a Bayesian mixture prior to perform the variable selection. Due to the binary nature of the data, the posterior distributions of the parameters are not in explicit form, and we need to use a combination of truncated sampling and Markov Chain Monte Carlo (MCMC) based computation techniques to simulate the posterior distributions. The Bayesian model is ?exible enough to identify the signi?cant genes as well as to perform future predictions. The method is applied to cancer classi?cation via cDNA microarrays. In particular, the genes BRCA1 and BRCA2 are associated with a hereditary disposition to breast cancer, and the method is used to identify the set of signi?cant genes to classify BRCA1 and others. Microarray data can also be applied to survival models. We address the issue of how to reduce the dimension in building model by selecting signi?cant genes as well as assessing the estimated survival curves. Additionally, we consider the wellknown Weibull regression and semiparametric proportional hazards (PH) models for survival analysis. With microarray data, we need to consider the case where the number of covariates p exceeds the number of samples n. Speci?cally, for a given vector of response values, which are times to event (death or censored times) and p gene expressions (covariates), we address the issue of how to reduce the dimension by selecting the responsible genes, which are controlling the survival time. This approach enables us to estimate the survival curve when n << p. In our approach, rather than ?xing the number of selected genes, we will assign a prior distribution to this number. The approach creates additional ?exibility by allowing the imposition of constraints, such as bounding the dimension via a prior, which in e?ect works as a penalty. To implement our methodology, we use a Markov Chain Monte Carlo (MCMC) method. We demonstrate the use of the methodology with (a) di?use large B??cell lymphoma (DLBCL) complementary DNA (cDNA) data and (b) Breast Carcinoma data. Lastly, we propose a mixture of Dirichlet process models using discrete wavelet transform for a curve clustering. In order to characterize these time??course gene expresssions, we consider them as trajectory functions of time and gene??speci?c parameters and obtain their wavelet coe?cients by a discrete wavelet transform. We then build cluster curves using a mixture of Dirichlet process priors.

Made available in DSpace on 2014-06-11T19:27:42Z (GMT). No. of bitstreams: 0 Previous issue date: 2007-01-30Bitstream added on 2014-06-13T19:35:45Z : No. of bitstreams: 1 haddad_sr_me_botfmvz.pdf: 305893 bytes, checksum: 02ff7e2de8b8a9ad8c10e7f22f207754 (MD5)
Conselho Nacional de Desenvolvimento Científico e Tecnológico (CNPq)
Universidade Estadual Paulista (UNESP)
O presente trabalho objetivou comparar, utilizando dados de um experimento de Microarray com um delineamento simples, os resultados de diferentes testes estatísticos a fim de verificar suas características na detecção de diferenças no nível de expressão dos genes. Os dados foram provenientes da South Dakota State University-EUA, do Department of Biology and Microbiology, Department of Animal Science, onde toda a parte experimental foi realizada. O material biológico envolveu quatro aves infectadas e quatro não infectadas com o vírus de bronquite infecciosa (IBV). O RNA utilizado foi extraído da camada epitelial da traquéia de animais controle e infectados com o vírus da IBV e, após a transcrição reversa foi marcado com os corantes fluorescentes (Cy3 e Cy5) e hibridizados com o microarray 13k cDNA de aves (FHCRC, Seattle, WA). A análise de dados dos resultados do experimento de microarray englobou dois estágios, sendo o primeiro denominado de Normalização, em que os dados foram pré-processados utilizando o procedimento Loess. A seguir foram realizadas as análises estatísticas propriamente ditas com testes de significância. Utilizou-se um modelo simples de ANOVA e aplicaram-se diferentes metodologias de análise. A análise das imagens revelou que dos 16192 spots em cada slide, apenas 10.926 puderam ser lidos sem defeitos no primeiro slide, 11.633 no segundo slide, 12577 no terceiro e 13.154 no quarto slide. A grande maioria dos spots em branco e controles negativos apresentou defeitos que determinaram sua eliminação. Um total de 13.597 spots foi lido no conjunto dos quatro slides, mas apenas 9.853 spots estavam representados em todos os slides. Concluiu-se que os experimentos de microarray, por tratarem de um conjunto muito grande de observações a serem analisados requerem análises estatísticas específicas. O método de Cui et al. (2005) reduziu...
The aim of this research was to compare, using real data of an experiment of Microarray with a simple design, the results of different statistical tests in order to verify their characteristics in the detection of differences in the level of expression of the genes. The data were coming of South Dakota State University-EUA, of the Department of Biology and Microbiology, Department Animal of Science, where the whole experimental part was accomplished. The biological material involved four infected animals and four no infected with the virus of infectious bronchitis (IBV). Used RNA was extracted of the layer epitelial of the windpipe of animals control and infected with the virus of IBV and, after the reverse transcription it was marked with the fluorescent colors (Cy3 and Cy5) and hybridization with the microarray 13k cDNA of birds (FHCRC, Seattle, WA). The analysis of data of the results of the microarray experiment included two apprenticeships, being the first denominated of Normalization, in that the data were pre-processed using the procedure Loess. To follow the statistical analyses they were accomplished properly said through real data with significant tests. A simple model of ANOVA was used and different analysis methodologies were applied. The analysis of the images revealed that of the 16192 spots in each slide, only 10.926 could be read without defects in the first slide, 11.633 in the second slide, 12577 in the third slide and 13.154 in the fourth slide. The great majority of the spots in white and negative controls presented defects that determined it elimination. A total of 13.597 spots was read in the group of the four slides, but only 9.853 spots were represented in all of the slides. It was ended that the microarray experiments, for they treat of a very big group of observations to be analyzed request specific statistical analyses. The method of Cui et al. (2005) it reduced... (Complete abstract click electronic access below)

Genome-wide microarray technology has facilitated the systematic discovery of diagnostic biomarkers of cancers and other pathologies. However, meta-analyses of published arrays using melanoma as a test cancer has uncovered significant inconsistences that hinder advances in clinical practice. In this study a computational model for the integrated analysis of microarray datasets is proposed in order to provide a robust ranking of genes in terms of their relative significance; both genome-wide relative significance (GWRS) and genome-wide global significance (GWGS). When applied to five melanoma microarray datasets published between 2000 and 2011, a new 12-gene diagnostic biomarker signature for melanoma was defined (i.e., EGFR, FGFR2, FGFR3, IL8, PTPRF, TNC, CXCL13, COL11A1, CHP2, SHC4, PPP2R2C, and WNT4). Of these, CXCL13, COL11A1, PTPRF and SHC4 are components of the MAPK pathway and were validated by immunocyto- and immunohisto-chemistry. These proteins were found to be overexpressed in metastatic and primary melanoma cells in vitro and in melanoma tissue in situ compared to melanocytes cultured from healthy skin epidermis and normal healthy human skin. One challenge for the integrated analysis of microarray data is that the microarray data are produced using different platforms and bio-samples, e.g. including both cell line- and biopsy-based microarray datasets. In order to address these challenges, the computational model was further enhanced the stratification of datasets into either biopsy or cell line derived datasets, and via the weighting of microarray data based on quality criteria of data. The methods enhancement was applied to 14 microarray datasets of three cancers (breast, prostate, and melanoma) based on classification accuracy and on the capability to identify predictive biomarkers. Four novel measures for evaluating the capability to identify predictive biomarkers are proposed: (1) classifying independent testing data using wrapper feature selection with machine leaning, (2) assessing the number of common genes with the genes retrieved in independent testing data, (3) assessing the number of common genes with the genes retrieved in across multiple training datasets, (4) assessing the number of common genes with the genes validated in the literature. This enhancement of computational approach (i) achieved reliable classification performance across multiple datasets, (ii) recognized more significant genes into the top-ranked genes as compared to the genes detected by the independent test data, and (iii) detected more meaningful genes than were validated in previous melanoma studies in the literature.

INTODUÇÃO: O carcinoma renal (CR) é uma doença agressiva, e sua incidência vem aumentando. A variante de células claras (CRCC) é a mais comum e apresenta comportamento biológico mais agressivo. Os recentes avanços no conhecimento da biologia molecular do tumor demonstram que a oncogênese dos diversos tipos histológicos é regida por mecanismos celulares diversos. Os modelos prognósticos atuais vêm procurando incorporar os recentes avanços da biologia molecular, com o intuito de melhorar sua capacidade de predizer a evolução e o desfecho destes pacientes. OBJETIVOS: Correlacionar a imunoexpressão dos marcadores selecionados com: 1) sobrevida global e, 2) com parâmetros prognósticos estabelecidos (estadio clínico TNM, tamanho tumoral, grau nuclear de Fuhrman, invasão microvascular e invasão de gordura perirrenal) em portadores de CRCC não metastático. MÉTODOS: Neste estudo de coorte retrospectivo, avaliamos 99 pacientes portadores de CRCC não metastático, quanto à expressão imunoistoquímica das seguintes proteínas: CA-IX, EGF-R, Ki-67, p53, PTEN, VEGF e VEGF-R. Os parâmetros analisados foram: Sobrevida global, estadio TNM, tamanho tumoral, grau nuclear de Fuhrman, invasão microvascular e invasão de gordura perirrenal. Utilizamos um tissue microarray construído exclusivamente para esta finalidade e realizamos a leitura da imunoexpressão por técnica digital utilizando o software Photoshop®. RESULTADOS: O tempo de seguimento médio foi de 7,9 anos. Com relação à sobrevida global, não observamos sua correlação com nenhum dos marcadores avaliados. Quanto à correlação da expressão dos marcadores com os parâmetros prognósticos convencionais, observamos que a expressão do EGF-R se correlacionou com estadio T (p= 0,049) e invasão da gordura perirrenal (p=0,020); e o VEGF-R se correlacionou com grau de Fuhrman (p=0,022) e invasão microvascular (p=0,005). Nos demais marcadores, não foi observada correlação significativa. CONCLUSÃO: Os fatores prognósticos moleculares EGF-R e VEGF-R apresentam-se como ferramentas úteis para avaliação do risco de prognóstico desfavorável em portadores de carcinoma renal de células claras não metastático
INTODUCTION: Renal cell carcinoma is an aggressive disease and its incidence is rising. The clear cell variant is the most common, and also the most aggressive. Recent advances in the understanding of the tumors molecular biology indicate that the oncogenesis of each histologic subtype is controlled by distinct cellular mechanisms. Current prognostic models are gradually incorporating the advances in molecular biology, in the hope to improve their predictive capacity. OBJECTIVES: To correlate the immunoexpression of selected markers with 1) overall survival, and 2) with established prognostic parameters (clinical TNM stage, tumor size, Fuhrman nuclear grade, microvascular invasion and perirenal fat invasion) in patients with non-metastatic ccRCC. METHODS: This is a retrospective cohort study, we evaluated 99 patients with non-metastatic clear cell renal cell carcinoma, as to the expression of the following proteins: CA-IX, EGF-R, Ki-67, p53, PTEN, VEGF e VEGF-R. The analyzed parameters where: overall survival, TNM stage, tumor size, Fuhrman nuclear grade, microvascular invasion, perirenal fat invasion. We utilized a custom built tissue microarray, and the immunoexpression was digitally quantified using the Photoshop® software. RESULTS: The mean follow-up time was 7,9 years. We found no correlation between the expression of the studied molecular markers and overall survival. As for the conventional prognostic parameters, we found the expression of EGF-R to correlate with T stage (p= 0,049) and perirenal fat invasion (p= 0,020), and VEGF-R to correlate with Fuhrman nuclear grade (p= 0,022) and microvascular invasion (p= 0,022). None of the other markers showed correlation with the studied parameters. CONCLUSIONS: The expression of EGF-R and VEGF-R may be useful tools in the prognostic evaluation of unfavorable risk in patients with non metastatic clear cell renal cell carcinoma

Clustering techniques are important for gene expression data analysis. However, efficient computational algorithms for clustering time-series data are still lacking. This work documents two improvements on an existing profile-based greedy algorithm for short time-series data; the first one is implementation of a scaling method on the pre-processing of the raw data to handle some extreme cases; the second improvement is modifying the strategy to generate better clusters. Simulation data and real microarray data were used to evaluate these improvements; this approach could efficiently generate more accurate clusters. A new feature-based algorithm was also developed in which steady state value; overshoot, rise time, settling time and peak time are generated by the 2nd order control system for the clustering purpose. This feature-based approach is much faster and more accurate than the existing profile-based algorithm for long time-series data.

Orientador: Gonçalo Amarante Guimães Pereira
Dissertação (mestrado) - Universidade Estadual de Campinas, Instituto de Biologia
Made available in DSpace on 2018-08-12T09:23:56Z (GMT). No. of bitstreams: 1 Lepikson-Neto_Jorge_M.pdf: 1765356 bytes, checksum: c73289c953d58c6b221c2e0927805499 (MD5) Previous issue date: 2008
Resumo: Com o intuito de obter informações relevantes para o melhoramento genético do eucalipto para a produção de biomassa, o presente trabalho buscou comparar a expressão dos genes relacionados com a formação e desenvolvimento da madeira em quatro diferentes espécies de eucalipto, tendo como objetivo identificar os padrões que as tornam mais aptas, bem como quais genes relacionados com determinadas características. Essa análise abre a possibilidade da identificação de genes chave que possam ser manipulados, através do melhoramento clássico ou da transgênia, para aumentar o conteúdo relativo de celulose das plantas, incrementando a sua eficiência para processos econômicos. 384 ESTs do banco de dados do Consórcio Genolyptus foram selecionadas para serem analisadas através da tecnologia de microarranjos de cDNA. Foram selecionadas ESTs de genes com funções conhecidas relacionadas com a formação da madeira, bem como de genes relacionados com o desenvolvimento do vegetal e de genes com função ainda desconhecida. Os dados obtidos foram cruzados com a biblioteca de ESTs do Consorcio Genolyptus (Northern Eletrônico), e foram feitos PCR em tempo real para os principais genes diferenciais nos microarranjos e para os genes da via de lignina e flavonóides. Os resultados mostraram que diferentes genes estão expressos nas espécies estudadas sendo um grande número deles ainda com função desconhecida no metabolismo do eucalipto. A maioria dos genes relacionados com a formação da parede celular não apresentou perfil de expressão diferencial nos microarranjos, sugerindo que as diferenças fenotípicas entre as madeiras das espécies estudadas podem estar sustentadas em vias alternativas, com fatores de elongação, cliclínas e outros genes desempenhando papel importante, bem como genes ainda não relacionados com o desenvolvimento da parede celular e ou ao desenvolvimento do vegetal. Os experimentos com PCR em tempo real mostraram que genes da via de flavonóides relacionados com a via de lignina e formação da madeira podem estar desempenhando papeis importantes no controle da formação da madeira da espécie. Esses resultados representam avanços significativos no entedimento da formação da madeira em eucalipto e servem como base para orientar futuras investigações no intuito de melhorar geneticamente esta espécie.
Abstract: In this work we intended to assemble relevant information for the genetic engineering of Eucalyptus for biomass production by comparing gene expression of genes related with the xylem and wood development of four different Eucalyptus species with distinct caractheristics, as a form to identify patterns and wich genes are possibly related to their differences. This analysis opens the possibilities to manipulate the specie and increase the overall celluloses content and its purpose for industrial production. 384 ESTs were selected from de Genolyptus database and analysed by microarryay cDNA experiments. Among the ESTs selected some were related to wood formation, cell wall assembly and some still had no general function known on the specie. The data from the microarray experiments were then crossed with the Genolyptus ESTs lybrary (Eletronic Northern) and Real-Time PCR were performed for the most relevant resultas as well as the genes from de lignin and flavonoid pathway. Results show that different genes are expressed among the xylem of the four species studied and most of them still have no related function to the metabolism of the plant. Most of the genes related to cell wall formation were not differentially expressed on the microarrays suggesting that the differences on the quality and structure of the wood among the four species might as well be resulted from the expression of alternative pathways and genes such as elongation factors, ciclins and others not yet related to the cell wall formation and wood development. Real-Time PCR experiments shown that genes from the flavonoid pathway related to lignin and wood formation might be playing a crucial role determining wicht pathway must be followed and therefore the type and quality of the wood on Eucalyptus. These results represent significant advances to our understanding on the formation of wood on Eucalyptus and will be valuable as a basis for future investigation aiming genetic engeneering of the specie.
Mestrado
Bioquimica
Mestre em Biologia Funcional e Molecular

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.

Orientador: Henrique Nunes de Oliveira
Banca: Raysildo Barbosa Lobo
Banca: Danísio Prado Munari
Resumo: O presente trabalho objetivou comparar, utilizando dados de um experimento de Microarray com um delineamento simples, os resultados de diferentes testes estatísticos a fim de verificar suas características na detecção de diferenças no nível de expressão dos genes. Os dados foram provenientes da South Dakota State University-EUA, do Department of Biology and Microbiology, Department of Animal Science, onde toda a parte experimental foi realizada. O material biológico envolveu quatro aves infectadas e quatro não infectadas com o vírus de bronquite infecciosa (IBV). O RNA utilizado foi extraído da camada epitelial da traquéia de animais controle e infectados com o vírus da IBV e, após a transcrição reversa foi marcado com os corantes fluorescentes (Cy3 e Cy5) e hibridizados com o microarray 13k cDNA de aves (FHCRC, Seattle, WA). A análise de dados dos resultados do experimento de microarray englobou dois estágios, sendo o primeiro denominado de Normalização, em que os dados foram pré-processados utilizando o procedimento Loess. A seguir foram realizadas as análises estatísticas propriamente ditas com testes de significância. Utilizou-se um modelo simples de ANOVA e aplicaram-se diferentes metodologias de análise. A análise das imagens revelou que dos 16192 spots em cada slide, apenas 10.926 puderam ser lidos sem defeitos no primeiro slide, 11.633 no segundo slide, 12577 no terceiro e 13.154 no quarto slide. A grande maioria dos spots em branco e controles negativos apresentou defeitos que determinaram sua eliminação. Um total de 13.597 spots foi lido no conjunto dos quatro slides, mas apenas 9.853 spots estavam representados em todos os slides. Concluiu-se que os experimentos de microarray, por tratarem de um conjunto muito grande de observações a serem analisados requerem análises estatísticas específicas. O método de Cui et al. (2005) reduziu... (Resumo completo, clicar acesso eletrônico abaixo)
Abstract: The aim of this research was to compare, using real data of an experiment of Microarray with a simple design, the results of different statistical tests in order to verify their characteristics in the detection of differences in the level of expression of the genes. The data were coming of South Dakota State University-EUA, of the Department of Biology and Microbiology, Department Animal of Science, where the whole experimental part was accomplished. The biological material involved four infected animals and four no infected with the virus of infectious bronchitis (IBV). Used RNA was extracted of the layer epitelial of the windpipe of animals control and infected with the virus of IBV and, after the reverse transcription it was marked with the fluorescent colors (Cy3 and Cy5) and hybridization with the microarray 13k cDNA of birds (FHCRC, Seattle, WA). The analysis of data of the results of the microarray experiment included two apprenticeships, being the first denominated of Normalization, in that the data were pre-processed using the procedure Loess. To follow the statistical analyses they were accomplished properly said through real data with significant tests. A simple model of ANOVA was used and different analysis methodologies were applied. The analysis of the images revealed that of the 16192 spots in each slide, only 10.926 could be read without defects in the first slide, 11.633 in the second slide, 12577 in the third slide and 13.154 in the fourth slide. The great majority of the spots in white and negative controls presented defects that determined it elimination. A total of 13.597 spots was read in the group of the four slides, but only 9.853 spots were represented in all of the slides. It was ended that the microarray experiments, for they treat of a very big group of observations to be analyzed request specific statistical analyses. The method of Cui et al. (2005) it reduced... (Complete abstract click electronic access below)
Mestre

This dissertation focuses on methodology specific to microarray data analyses that organize the data in preliminary steps and proposes a cluster analysis method which improves the interpretability of the cluster results. Cluster analysis of microarray data allows samples with similar gene expression values to be discovered and may serve as a useful diagnostic tool. Since microarray data is inherently noisy, data preprocessing steps including smoothing and filtering are discussed. Comparing the results of different clustering methods is complicated by the arbitrariness of the cluster labels. Methods for re-labeling clusters to assess the agreement between the results of different clustering techniques are proposed. Microarray data involve large numbers of observations and generally present as arrays of light intensity values reflecting the degree of activity of the genes. These measurements are often two dimensional in nature since each is associated with an individual sample (cell line) and gene. The usual hierarchical clustering techniques do not easily adapt to this type of problem. These techniques allow only one dimension of the data to be clustered at a time and lose information due to the collapsing of the data in the opposite dimension. A novel clustering technique based on normal mixture distribution models is developed. This method clusters observations that arise from the same normal distribution and allows the data to be simultaneously clustered in two dimensions. The model is fitted using the Expectation/Maximization (EM) algorithm. For every cluster, the posterior probability that an observation belongs to that cluster is calculated. These probabilities allow the analyst to control the cluster assignments, including the use of overlapping clusters. A user friendly program, 2-DCluster, was written to support these methods. This program was written for Microsoft Windows 2000 and XP systems and supports one and two dimensional clustering. The program and sample applications are available at http://etd.vcu.edu. An electronic copy of this dissertation is available at the same address.

Poverty and long distances are two reasons why some people in the third world countries hasdifficulties seeking medical help. A solution to the long distances could be if the medical carewas more mobile and diagnostically tests could be performed on site in villages. A new pointof-care test based on a small blood shows promising results both in run time and mobility.However, the method still needs more advanced equipment for analysis of the resultingmicroarray. This study has investigated the potential to perform the analysis within asmartphone application, performing all steps from image capturing to a diagnostic result. Theproject was approach in two steps, starting with implementation and selection of imageanalysis methods and finishing with implementing those results into an Android application.A final application was not developed, but the results gained from this project indicates that asmartphone processing power is enough to perform heavy image analysis within a sufficientamount of time. It also imply that the resolution in the evaluated images taken with a Nexus 6together with an external macro lens most likely is enough for the whole analysis, but furtherwork must be done to ensure it.

The high-throughput technologies developed in the last decade have revolutionized the speed of data accumulation in the life sciences. As a result we have very rich and complex data that holds great promise to solving many complex biological questions. One such technology that is very well established and widespread is DNA microarrays, which allows one to simultaneously measure the expression levels of tens of thousands of genes in a biological tissue. This thesis aims to contribute to the development of statistics that allow the end users to obtain robust and meaningful results from DNA microarrays for further investigation. The methodology, implementation and pragmatic issues of two important and related topics – sample size estimations for designing new studies and meta-analysis of existing studies – are presented here to achieve this aim. Real life case studies and guided steps are also given. Sample size estimation is important at the design stage to ensure a study has sufficient statistical power to address the stated objective given the financial constraints. The commonly used formula for estimating the number of biological samples, its short-comings and potential amelioration are discussed. The optimal number of biological samples and number of measurements per sample that minimizes the cost is also presented. Meta-analysis or the synthesis of information from existing studies is very attractive because it can increase the statistical power by making comprehensive and inexpensive use of available information. Furthermore, one can also easily test the generalizability of findings (i.e. the extent of results from a particular valid study can be applied to other circumstances). The key issues in conducting a meta-analysis for microarrays studies, a checklist and R codes are presented here. Finally, the poor availability of raw data in microarray studies is discussed here with recommendations for authors, journal editors and funding bodies. Good availability of data is important for meta-analysis in order to avoid biased results and for sample size estimation.

Les micropuces sont devenues incontournables pour la recherche post-génomique en permettant de cribler plusieurs milliers de composés dans quelques microlitres. Plusieurs stratégies ont été développées pour immobiliser sur ce format des petites molécules ou des anticorps pour la recherche de médicaments ou comme diagnostique. L’encodage de chimiothèque de petites molécules par des ‘Peptid Nucleic Acid‘ (PNA), possible par synthèse combinatoire (split and mix), permet d’organiser des micropuces par auto-assemblage des sondes sur ce support et de cribler ainsi l’activité enzymatique sans a priori à partir de mélange complexe tel que des lysats cellulaire. Nous avons développé la chimie peptidique et des PNA pour synthétiser et cribler 3 générations de chimiothèques d’inhibiteurs encodé par des PNA. Ces chimiothèques présentent 625 ou 4000 inhibiteurs basés sur le mécanisme d’action de protéases, étiqueté par une séquence unique de PNA encodant la structure d’un inhibiteur tetrapeptidiques ainsi que sa localisation spécifique sur une micropuce. De gros efforts portés sur l’optimisation du dépôt sur micropuce, la détection du signal et les conditions d’hybridation ont rendu cette stratégie suffisamment sensible et spécifique pour quantifier l’activité enzymatique et pour identifier des spécificités de substrats. Cette technique a été utilisée pour profiler l’activité enzymatique d’extrait allergisant d’acariens et pour détecter des spécificités de substrats d’enzymes purifiés. Les inhibiteurs identifiés sont assez spécifique pour distinguer l’activité d’enzymes proches d’une même famille et les enzymes cibles de ces inhibiteurs peuvent être identifiés par colonne d’affinité couplé à la spectrométrie de masse. De plus, les inhibiteurs identifiés permettent de bloquer l’activité d’un enzyme et d’évaluer sa relation avec un phénotype. Cette stratégie est actuellement la seule méthode fonctionnelle miniaturisée permettant de profiler l’activité enzymatique avec des inhibiteurs
Microarray has become an indispensable technology in postgenomic research area and allows for high throughput screening of several thousand analytes in few microliters. Several strategies have been developed to immobilize small molecules or antibodies on this format for drug discovery or for diagnostic tool. The peptide nucleic acid (PNA)-encoded small molecule library strategy allows for combinatorial libraries synthesize in a split and mix format to be organized into microarray by a self-sorting assembly. This methodology enables enzyme activity screening without a priori knowledge of the target in complex mixture such as crude cell lysates. Peptide and PNA chemistries were developed to synthesize and screen on a microarray format 3 generations of PNA-encoded inhibitor libraries targeting several protease families. Those libraries are presenting 625 to 4000 mechanism based inhibitors individually labelled with a PNA sequences that encodes the structure of a tetrapeptide inhibitor and its specific localisation onto a microarray. Substantial efforts on optimizing microarray spotting as well as signal detection and hybridization conditions presented in this thesis gives to PNA-encoded strategy a sensitive and specific format to quantify enzymatic activities and identify substrate specificities. The PNA-encoded strategy has been used to profiling enzyme activity from allergenic dustmite extract as well as to detect substrate specificities of purified enzyme samples. Identified inhibitors showed to be specific enough to decipher closely related activities of enzyme from the same family, and on the other hand, to identify the inhibitor’s target enzyme by affinity column coupled to mass spectrometry. Finally, identified inhibitors can be used to knock down the activity of an enzyme and evaluate its correlation to a phenotype. This strategy is to date the only functional methodology that enable to profile enzyme activity with inhibitors in a miniaturized format

Statisticians are used to model interactions between variables on the basis of observed

data. In a lot of emerging fields, like bioinformatics, they are confronted with datasets

having thousands of variables, a lot of noise, non-linear dependencies and, only, tens of

samples. The detection of functional relationships, when such uncertainty is contained in

data, constitutes a major challenge.

Our work focuses on variable selection and network inference from datasets having

many variables and few samples (high variable-to-sample ratio), such as microarray data.

Variable selection is the topic of machine learning whose objective is to select, among a

set of input variables, those that lead to the best predictive model. The application of

variable selection methods to gene expression data allows, for example, to improve cancer

diagnosis and prognosis by identifying a new molecular signature of the disease. Network

inference consists in representing the dependencies between the variables of a dataset by

a graph. Hence, when applied to microarray data, network inference can reverse-engineer

the transcriptional regulatory network of cell in view of discovering new drug targets to

cure diseases.

In this work, two original tools are proposed MASSIVE (Matrix of Average Sub-Subset

Information for Variable Elimination) a new method of feature selection and MRNET (Minimum

Redundancy NETwork), a new algorithm of network inference. Both tools rely on

the computation of mutual information, an information-theoretic measure of dependency.

More precisely, MASSIVE and MRNET use approximations of the mutual information

between a subset of variables and a target variable based on combinations of mutual informations

between sub-subsets of variables and the target. The used approximations allow

to estimate a series of low variate densities instead of one large multivariate density. Low

variate densities are well-suited for dealing with high variable-to-sample ratio datasets,

since they are rather cheap in terms of computational cost and they do not require a large

amount of samples in order to be estimated accurately. Numerous experimental results

show the competitiveness of these new approaches. Finally, our thesis has led to a freely

available source code of MASSIVE and an open-source R and Bioconductor package of

network inference.
Doctorat en sciences, Spécialisation Informatique
info:eu-repo/semantics/nonPublished

Previous studies have demonstrated specific biochemical abnormalities in the noradrenergic locus coeruleus (LC) that are strongly associated with major depressive disorder (MDD). Here, we studied the LC of 4 pairs of MDD and matched control subjects by gene expression microarray analysis in an effort to accelerate the discovery of pathobiological abnormalities of these cells in MDD. Among matching criteria, pH values of control (6.71±0.06) and MDD (6.66±0.12) subjects were closely matched. Gene expression profiling using whole human genome microarrays (Agilent) revealed statistically significant changes in approximately 50 transcripts in the LC of depressive subjects. Quantitative real-time PCR (qPCR) was used to analyze transcripts identified by microarray anlayses. In initial studies of 11 of these transcripts that demonstrated a >2-fold change in microarrays, only 3 transcripts were confirmed by qQPCR in a larger sample of 11-12 pairs of MDD and matched control subjects. Amounts of bone morphogenetic factor-7 (BMP7; p=0.001) and potassium channel subfamily K, member 7 (KCNK7; p=0.049) mRNAs were significantly lower in MDD subjects compared to control subjects (~2-fold difference). In contrast, neurolysin mRNA levels were significantly higher (~3-fold; p=0.03) in MDD than in control subjects. BMP7 is a member of the TGF-β superfamily and has neuroprotective and neurotrophic effects on catecholaminergic neurons. The KCNK family of potassium channels contribute to the excitability of neurons. Neurolysin is a zinc-dependent metallopeptidase involved in neuropeptide metabolism. The present study is the first report of these novel gene expression abnormalities in the LC of MDD subjects. These findings enhance our understanding of the pathobiology of MDD and may represent novel targets for pharmacological management of depression.