INTRODUÇÃO: A cirurgia associada à radioterapia proporciona altas taxas de preservação de membros e de controle local em sarcomas de partes moles de extremidade de adultos, mas ainda apresenta elevadas taxas de complicações locais e de metástases à distância. O valor da quimioterapia adjuvante ou neoadjuvante ainda é controverso e objeto de investigações clínicas. A identificação de fatores moleculares preditivos de resposta à quimioterapia pode selecionar pacientes que se beneficiem ou não da sua aplicação. OBJETIVOS: identificar perfis de expressão gênica capazes de diferenciar tumores respondedores e não respondedores a quimioterapia neoadjuvante em sarcomas de partes moles. Analisar os resultados preliminares relativos à efetividade de um esquema de quimioterapia neoadjuvante em sarcomas de partes moles. MÉTODOS: amostras foram coletadas a partir de um ensaio clínico fase II não controlado que testa um esquema de quimioterapia neoadjuvante com doxorrubicina e ifosfamida em sarcomas de alto grau histológico, localizados em extremidades de pacientes adultos. O perfil de expressão gênica foi determinado pela análise de cDNA microarrays. RESULTADOS: 14 pacientes foram incluídos no estudo clínico e 6 amostras foram utilizadas para análise molecular. 222 seqüências diferentemente expressas entre respondedores e não respondedores foram identificadas. Entre os genes com maior diferença de expressão, foram observados genes envolvidos com via de sinalização de TGF, genes envolvidos com angiogênese, com degradação de matriz extracelular e com desenvolvimento. A taxa de resposta objetiva à quimioterapia neoadjuvante foi de 28,6%, a taxa de amputação foi de 7,1% e taxa de complicações relacionadas à ferida operatória foi de 23%. Complicações graus 3 e 4 ocorreram em 50% dos pacientes e nenhum deles faleceu ou teve a proposta cirúrgica suspensa em decorrência de complicações da quimioterapia. CONCLUSÕES: tumores respondedores a quimioterapia neoadjuvante com doxorrubicina e ifosfamida apresentaram um perfil de expressão gênica diferente dos não respondedores, particularmente em genes envolvidos na via de sinalização de TGF. O esquema terapêutico testado mostrou-se efetivo e seguro para ser investigado em um estudo fase III / INTRODUCTION: Surgery combined with adjuvant radiotherapy provides high rates of limb sparing and local control for adult extremity soft tissue sarcomas, but is still associated with high rates of local morbidity and distant recurrences. The role of adjuvant or neoadjuvant chemotherapy is still controversy and target of clinical investigations. The identification of molecular predictive factors of response to chemotherapy could select patients who have benefits or not with its use. OBJECTIVES: to identify gene expression profiles that discriminate tumors with respect to response to neoadjuvant chemotherapy. Analyze the preliminary results of a protocol of neoadjuvant chemotherapy in soft tissue sarcomas. METHODS: samples were collected from subjects of a single-arm prospective clinical trial that investigates the effectiveness of a neoadjuvant doxorubicin and ifosphamide-based chemotherapy regimen in high grade extremity soft tissue sarcomas in adults. Gene expression profiles were determined by the analysis of cDNA microarrays. RESULTS: 14 patients were included in the clinical trial and six samples were used in the molecular study. 222 sequences differentially expressed between responders and non responders were identified. Among the genes with higher differences in expression, we have identified genes involved with TGF signaling pathway, angiogenesis, extracelular matrix degradation and development. The objective response rate to neoadjuvant chemotherapy was 28,6%, the amputation rate was 7,1%, and the wound complication rate was 23%. Grades 3 and 4 complications have occurred in 50 % of the cases, but no deaths or modifications on surgical intent related to chemotherapy complications have occurred. CONCLUSIONS: tumors considered responders to neoadjuvant chemotherapy showed a gene expression profile significantly different from non responders, especially with respect to the TGF signaling pathway. The neoadjuvant regimen tested has showed to be effective and safe to be considering for a phase III clinical trial

INTRODUÇÃO: O papilomavírus humano (HPV) encontra-se amplamente distribuído na população mundial. Apesar da grande maioria das infecções serem transientes, assintomáticas e passíveis de regressão espontânea, a infecção persistente por tipos de alto risco de HPV é necessária para o desenvolvimento de neoplasias intraepiteliais cervicais. Uma vez que apenas uma pequena parcela das infecções progride à lesões malignas após um longo período desde o diagnóstico inicial de lesões precursoras, tem-se iniciado a busca por fatores que possam influenciar na progressão ou na eliminação destas manifestações iniciais. A variabilidade genética viral tem sido apontada como um dos fatores que interagem neste processo. Embora virtualmente todos os tumores da cérvice uterina apresentem o DNA viral, neoplasias em outros sítios anatômicos têm sido apenas em parte correlacionadas com a presença viral, sendo o HPV proposto como um dos agentes causadores de tumores em sítios de cabeça e pecoço. MÉTODOS: Espécimens clínicos de tumores de cabeça e pescoço, fixados em formalina e contidos em parafina (FFPE), provenientes do Instituto do Câncer do Estado de São Paulo (n=79) e da Santa Casa de Misericórida de São Paulo (n=94), tiveram seu DNA extraído, seguido de diagnóstico e genotipagem de HPV pela metodologia de Inno-LiPA. Análises de linhagens moleculares foram realizadas nas amostras HPV-16 positivas. Análise imunohistoquímica de P16INK4a foi realizada em todas as amostras. RESULTADOS: A presença do DNA viral foi encontrada em 24,1% (19/79) dentre a série de tumores provenientes do ICESP, sendo a cavidade oral o sítio em que foi observada a maior proporção de DNA viral (27,1%), enquanto que 13,8% (13/94) dentre os espécimens provenientes da Santa Casa apresentaram-se positivos para HPV, sendo a cavidade oral o sítio em que foi observada a maior proporção do DNA viral (18,1%). O HPV-16 foi o tipo mais prevalente, detectado em 73,4% das amostras HPV positivas provenientes do ICESP e 61,5% das amostras provenientes da Santa Casa. Independente da Instituição, as amostras foram alocadas no clado das linhagens Asiático-Americana e Europeia em 50%, cada uma, entre os 18 tumores HPV-16 positivos em que as análises de linhagem foram possíveis. Não foi observada, nestas séries, correlação entre a superexpressão de P16INK4a e a presença do DNA viral. CONCLUSÃO: Nas amostras analisadas, o DNA de HPV foi detectado em 18,5% dos 173 espécimens. O HPV-16 foi o tipo mais prevalente. Isolados da linhagem Europeia e da linhagem Asiatico-Americano foram detectados em 50% dos casos, cada uma, dentre as amostras HPV-16 analisadas por este estudo / INTRODUCTION: Human papillomaviruses (HPV) are widely distributed worldwide. Although the majority of infections are usually transient, asymptomatic and frequently regress spontaneously, persistent infections by high-risk HPVs are necessary for the development of cervical intraepithelial neoplasia. Once only a small proportion of infections progress to malignant lesions after a long period of time since the initial diagnosis of precursor lesions, the search for factors that might influence the progression or clearence of these early manifestations are currently under way. Viral genetic variability has been proposed as one of the factors interacting in this process. Although virtually all cervix tumors present the viral DNA, neoplasias from other anatomical sites have been only in part correlated with viral presence, and HPV has been proposed as one causative agent in tumors from head and neck sites. METHODS: Clinical specimens of formalin-fixed paraffin embedded head and neck tumors, provided by the Cancer Institute of São Paulo (n=79) and also by the Santa Casa de Misericórdia de São Paulo (n=94), were submitted to DNA extraction and further HPV diagnostic and genotyping by the Inno-LiPA methodology. Molecular lineages analyses were performed in all HPV-16 positive samples. P16INK4a immunohistochemical analyses were conducted in all samples. RESULTS: HPV DNA was detected among 24.1% (19/79) of samples provided by ICESP, tumors from oral cavity presented the highest viral positivity (27.1%), whereas 13,8% (13/94) of the samples from Santa Casa presented HPV DNA, tumors from the oral cavity also presented the highest HPV positivity with 18.1% of viral DNA presence. HPV-16 was the most prevalent type detected in 73.4% and 61.5% of HPV positive ICESP and Santa Casa samples, respectively. Irrespective of the Institution, samples submitted to lineage analyzes were allocated in the Asiatic-American and European phylogenetic branches in 50%, each one, among the 18 tumors HPV-16 positive for which lineage analysis was possible. No correlation between P16INK4a overexpression and HPV DNA presence was observed. CONCLUSION: In this study, HPV DNA was detected in 18.5% among 173 head and neck tumor specimens. HPV-16 was the most prevalent type. The European and the Asiatic-American lineage were detected in 50% of the cases, each one, among the cases HPV-16 positive analyzed

A síndrome de Bloom (SB) é uma síndrome de instabilidade cromossômica rara, transmitida por herança autossômica recessiva. Caracteriza-se por deficiência de crescimento pré e pós-natal, microcefalia, hipoplasia malar, eritema telangiectásico em face e comprometimento do sistema imunológico, entre outras manifestações clínicas. Os pacientes com SB apresentam predisposição aumentada para o desenvolvimento de neoplasias em idade precoce, sendo esta, a principal causa de óbito. Ao estudo citogenético observa-se aumento de quebras cromossômicas espontâneas e trocas entre cromátides irmãs (TCI), que são utilizadas como marcador diagnóstico para a SB. Além disso, a literatura mostra que a maioria dos pacientes também apresenta mutações no gene BLM, que estão relacionadas a defeitos no mecanismo de reparo do DNA. No entanto, os mecanismos fisiopatológicos não são completamente compreendidos. Nesse sentido estudamos o transcriptoma de duas pacientes portadoras da síndrome de Bloom e de três controles utilizando a metodologia RNA-seq (Illumina, Inc., San Diego, CA). A análise de expressão diferencial revelou 216 genes diferencialmente expressos relacionados a vias relacionadas à resposta imune como replicação negativa da regulação da replicação do genoma viral, regulação positiva da proliferação de células B, via de sinalização mediada por interferon gama, ativação de células B, resposta a vírus, resposta imune adaptativa e processo efetor imune, e nenhuma diferença da expressão em genes de reparo de DNA. Concomitantemente, observamos a hiperexpressão do gene BLM para ambas as pacientes, contribuindo para a desestabilização de genes envolvidos em vias imunológicas, fenômeno também observado em alguns tumores. Dessa forma, sugerimos que a combinação de defeitos de proliferação linfocitária e defeitos de sinalização celular somados a outros, como perda celular e expressão alterada do gene BLM, podem contribuir diretamente para as principais características observadas na síndrome de Bloom, como a deficiência de crescimento e o elevado risco de câncer. Futuramente, o estudo do transcriptoma, aplicado a outros portadores da SB e outras síndromes de instabilidade, possibilitará uma análise mais acurada das interações gênicas relevantes para a desestabilização do genoma / Bloom Syndrome (BS) is a rare chromosome instability syndrome, with recessive autosomal inheritance. The main clinical manifestations are pre and postnatal growth deficiency, microcephaly, malar hypoplasia, telangiectasic facial erythema and compromised immune system, among others. Patients with BS present increased risk to the development of neoplasias at an early age, which is the main cause of death. Cytogenetic test is used as a diagnostic marker for BS since the patient\'s cells present increase in spontaneous chromosomal breaks and sister chromatid exchange (SCE). In addition, the literature reveals that most patients also present mutations in the BLM gene, which are related to defects in the DNA repair mechanism; however, it is still not completely understood. In this sense, we studied the transcriptome of two patients with Bloom\'s syndrome and three controls using the RNA-seq methodology (Illumina, Inc., San Diego, CA). Differential expression analysis revealed 216 differentially expressed genes related to immunological pathways such as: negative replication of the regulation of the viral genome replication, positive regulation of B cells proliferation, gama-interferon mediated signalization pathway, B cells activation, virus response, adaptive immune response and immune effector process, and absence of difference of DNA repair genes expression. At the same time, we observed the hyperexpression of the BLM gene for both patients contributing for the destabilization of genes involved in immunological pathways, a phenomenon also observed in some tumors. Thus, we suggest that the combination of lymphoid proliferation defects and cell signaling defects added to others such as cell loss and altered expression of the BLM gene may contribute directly to the main characteristics observed in Bloom\'s syndrome, such as growth failure and high risk of cancer. In the future, the study of the transcriptome applied to other BS carriers and other instability syndromes, will allow a more accurate analysis of the relevant gene interactions to the destabilization of the genome

Submitted by Renata Lopes (renatasil82@gmail.com) on 2017-06-21T13:51:59Z No. of bitstreams: 1 vivianhonoratobarletta.pdf: 2632443 bytes, checksum: c92921b16b55178bf7a44b922bd843ef (MD5) / Approved for entry into archive by Adriana Oliveira (adriana.oliveira@ufjf.edu.br) on 2017-08-07T19:10:07Z (GMT) No. of bitstreams: 1 vivianhonoratobarletta.pdf: 2632443 bytes, checksum: c92921b16b55178bf7a44b922bd843ef (MD5) / Made available in DSpace on 2017-08-07T19:10:07Z (GMT). No. of bitstreams: 1 vivianhonoratobarletta.pdf: 2632443 bytes, checksum: c92921b16b55178bf7a44b922bd843ef (MD5) Previous issue date: 2011-02-24 / CAPES - Coordenação de Aperfeiçoamento de Pessoal de Nível Superior / Os norovírus (NoV) são importantes agentes etiológicos, responsáveis por surtos e casos esporádicos de doença diarréica aguda, que acometem indivíduos de todas as idades. A partícula viral não apresenta envelope e o material genético é constituído por uma molécula de RNA de fita simples, de polaridade positiva. Pertencem à família Caliciviridae, gênero Norovirus e estão classificados em cinco genogrupos (GI–V), sendo que os NoV humanos estão agrupados nos genogrupos I, II e IV e destes, os NoV do genogrupo II e genótipo 4 (GII.4) são os mais comumente encontrados, em todo o mundo. Apesar da associação destes vírus com a doença diarréica aguda estar bem documentada na literatura mundial, no Brasil, os trabalhos são escassos e restritos aos grandes centros e adjacências. Assim, considerando-se o pouco conhecimento sobre os norovírus e a inexistência de dados epidemiológicos na cidade de Juiz de Fora, MG, foi realizado o presente estudo, cujos objetivos foram a detecção e caracterização molecular de amostras de NoV, associadas a casos esporádicos de doença diarréica aguda infantil, bem como a avaliação da influência de fatores climáticos e demográficos na ocorrência destas infecções. De janeiro de 2008 a dezembro de 2009, 218 espécimes fecais foram analisados para a presença de NoV, por RT-PCR convencional, todos obtidos de crianças de 0 a 12 anos de idade, proveniente de atendimentos ambulatoriais (89,45%) e hospitalizados (10,55%). Foram detectadas 20 (9,17%) amostras positivas e observou-se uma tendência de sazonalidade das infecções no período da estação seca, no ano de 2008, fato que não se repetiu em 2009. A maioria das amostras positivas foi detectada em crianças na faixa de 0 a 36 meses e não houve correlação, estatisticamente significante, entre a ocorrência das infecções e o sexo. Das 20 amostras detectadas, 19 foram caracterizadas como NoV GII e 1 como NoV GI. O sequenciamento parcial do genoma e a análise filogenética das amostras selecionadas, revelou a presença de NoV dos genótipos GII.4 e GII.6, que cocircularam nos dois anos do estudo. As amostras NoV GII.4, detectadas em Juiz de Fora, apresentaram maior similaridade de nucleotídeos e de aminoácidos com aquelas que circularam no estado do Rio de Janeiro nos anos de 2006, 2007 e 2008. A análise filogenética das amostras NoV GII.6 detectadas em Juiz de Fora, associada à alta similaridade das sequências de nucleotídeos e aminoácidos, mostrou que estas foram mais proximamente relacionadas com a amostra NoV GII.6 (GU132461/2007), detectada no estado do Rio de Janeiro em 2007, fatos que, aliados à proximidade geográfica de ambas as cidades, sugerem uma possível linhagem comum entre as mesmas. Este levantamento epidemiológico permitiu constatar a presença e circulação de NoV na população infantil de Juiz de Fora, MG, demonstrando sua importante participação como agente etiológico das diarreias agudas, também nesta comunidade / Noroviruses (NoV) are important etiological agents responsible for outbreaks and sporadic cases of acute diarrhea in individuals of all ages. The viral particles are nonenveloped with and the genome is composed of a positive single-stranded RNA. Norovirus belongs to the Caliciviridae family, Norovirus genus and are classified into five genogroups (GI-V), with GI, GII and GIV being found in human and among them, the NoV GII genotype 4 (GII.4) are the most commonly found worldwide. In Brazil, norovirus surveys are realized mainly in research institutes, carried out in the biggest centers and surroundings. Thus, considering the little knowledge about these viruses and the lack of epidemiological data on this viral infection in the Juiz de Fora city, MG state, it was performed this study, which aimed to detect and characterize the NoV samples, associated with sporadic cases of acute infantile diarrhea, as well as asses the influence of climatic and demographic factors in the occurrence of these infections. Between January 2008 to December 2009, 218 fecal specimens were analyzed for the presence of NoV by conventional RT-PCR, all obtained from children 0-12 years of age, from outpatient (89.45%) and inpatients (10.55%). We detected 20 (9.17%) positive samples and there was a tendency for seasonal infections during the dry season in 2008, a fact which was not repeated in 2009. The biggest number of positive samples were detected in children aged 0 to 24 months and there was no statistically significant correlation between the occurrence of infections and sex. Of the 20 samples detected, 19 were characterized as NoV GII and 1 as NoV GI. The partial genome sequencing and phylogenetic analysis of selected samples revealed the presence of NoV genotypes GII.4 and GII.6, which co-circulated in the two years of study. Samples NoV GII.4 detected in Juiz de Fora, showed greater similarity of nucleotides and aminoacids with those that circulated in the state of Rio de Janeiro during 2006, 2007 and 2008. Phylogenetic analysis of the samples GII.6 NoV, detected in Juiz de Fora, associated with the high similarity of nucleotide and amino acid sequences showed that they were most closely related to the sample GII.6 NoV (GU132461/2007) detected in the state of Rio de Janeiro in 2007. This fact associated with the geographical proximity of both cities, suggesting a possible common lineage between them. This epidemiological survey revealed the presence and circulation of NoV in the infantile population of Juiz de Fora, MG, demonstrating its important role as an etiologic agent of acute diarrhea, also in this community.

Orientadores: Anelyssa Cysne Frota D'Abreu, Iscia Teresinha Lopes-Cendes / Dissertação (mestrado) - Universidade Estadual de Campinas, Faculdade de Ciências Médicas / Made available in DSpace on 2018-08-25T05:40:31Z (GMT). No. of bitstreams: 1 Piovesana_LuizaGonzaga_M.pdf: 1656872 bytes, checksum: 2a3571bb0d9dc43e8f0405aa2c799a5e (MD5) Previous issue date: 2014 / Resumo: As distonias caracterizam-se por movimentos involuntários e torsionais, que se manifestam de diferentes formas e podem afetar quaisquer músculos voluntários. Diversas mutações genéticas foram associadas às distonias primárias, destacando-se a DYT1, DYT5 e DYT6. O gene DYT1/TOR1A foi o primeiro identificado, ao apresentar uma deleção GAG que produz uma proteína mutante que altera conexões núcleo-cito-esqueléticas e o processamento proteico. O fenótipo típico inicia-se dos três aos 26 anos, tem penetrância de 30%, sendo 60% com acometimento generalizado ou multifocal. O gene DYT6/THAP1 possui diversos polimorfismos descritos e acredita-se que seja o segundo em prevalência entre as distonias hereditárias. Possui penetrância ainda menor e fenótipo de início precoce, envolvimento crânio-cervical e de fala. A frequência, etiologia e as alterações genéticas das distonias não são conhecidas na população brasileira, que por suas características específicas, não podem ter os resultados de outras populações simplesmente transpostos para a nossa. Nosso objetivo, portanto, foi caracterizar genética e clinicamente uma amostra de pacientes brasileiros com distonia. Os indivíduos foram recrutados no Ambulatório de Distúrbios do Movimento e de Distonia. Os pacientes foram avaliados por um questionário padronizado, seguido por pesquisa das mutações DYT1 e DYT6. A avaliação clínica demonstrou que a nossa coorte apresenta padrão semelhante ao internacional, com pacientes de início jovem tendendo a apresentar quadros generalizados e com história familiar positiva e pacientes adultos mantendo quadros focais ou segmentares e esporádicos. A pesquisa de mutações identificou 1 mutação missense já descrita na literatura, envolvida na penetrância da DYT1 em indivíduos com a mutação causadora clássica da doença e que pode conferir risco quando encontrada isoladamente. Também encontramos uma deleção de 6 pares de base no fim do exon 1 do gene THAP1 que em avaliações preliminares altera o sitio de splicing e acaba por abortar a tradução da proteína por meio de um stop códon precoce. Concluímos que nossos pacientes tem apresentação clinica semelhante a literatura mundial, porém com características genotípicas diferentes, pois não encontramos em nenhum individuo as mutações mais classicamente associadas as doenças. A diferenciação das distonias em subtipos e o entendimento das vias moleculares comuns devem fazer parte de investigações futuras / Abstract: Dystonia is characterized by involuntary and torsional movements, which manifest themselves in various forms and can affect any voluntary muscle. Several genetic mutations have been associated with isolated hereditary dystonia, most notably the DYT1 and DYT6. The DYT1 is caused by a GAG deletion at the TOR1A gene, which produces a protein that alters intracellular membranes connections and protein processing. The typical phenotype starts from three to 26 years of age, it has penetrance of 30 % and 60% of the subjects will progress to a generalized or multifocal involvement. The DYT6/THAP1 gene has several polymorphisms described, believed to be the second in prevalence among the hereditary dystonias. DYT6 has an even lower penetrance. It is characteristically an early-onset, dystonia with craniocervical and speech involvement. The frequency, etiology and genetic alterations of dystonia are not known in the Brazilian population, which due to its specific characteristics may not share of the same genetic background. Our aim was to evaluate genetically and clinically a sample of Brazilian patients with dystonia. Subjects were recruited from the Outpatient Movement Disorders and Dystonia Clinics. Patients were assessed by a standardized questionnaire, followed by molecular testing for DYT1 and DYT6. Our cohort showed similar results with the literature, where young, generalized and positive family history commonly group together, as focal and segmental cases that appear in adulthood tend to stabilize. Sequencing of the DYT1 gene found a known missense mutation that contributes to low penetrance in individual with the typical DYT1 mutation but might be a risk factor when found in isolation. Sequencing of the DYT6 gene showed a novel 6bp deletion at the end of exon 1, causing alternate splicing and a premature stop codon in preliminary analysis. Subdividing phenotypes and understanding common molecular pathways through new genetic data is in the future of dystonia research / Mestrado / Fisiopatologia Médica / Mestra em Ciências

Orientadores: Carmen Silvia Passos Lima, Fernando Ferreira Costa / Tese (doutorado) - Universidade Estadual de Campinas, Faculdade de Ciências Médicas / Made available in DSpace on 2018-08-19T13:59:25Z (GMT). No. of bitstreams: 1 Lourenco_GustavoJacob_D.pdf: 4599367 bytes, checksum: 0a02bc7c2e5acd6b81b6cbb948444b69 (MD5) Previous issue date: 2011 / Resumo: Alterações genéticas herdadas, como os polimorfismos gênicos de base única (SNPs) e as variações no número de cópias do DNA (CNVs), foram associadas com o risco de carcinoma de células escamosas (CEC) de base de língua (BL) em poucos estudos. O CEC de BL é um tumor que determina altas taxas de morbidade e mortalidade, no entanto, sua associação com polimorfismos genéticos não está estabelecida. Frente ao exposto, o objetivo deste estudo foi o de avaliar os papéis de SNPs e CNVs no CEC de BL. O DNA genômico foi obtido de amostras de sangue periférico de 49 pacientes com CEC de BL e de 49 controles. Cada amostra foi analisada por meio de lâminas com microarranjos de DNA contendo 500.568 SNPs e 420.000 CNVs (Affymetrix®). A digestão enzimática do DNA, a ligação de adaptadores, a amplificação, a fragmentação, a marcação, a hibridização, as lavagens e a leitura das intensidades dos sinais das sondas foram realizadas de acordo com instruções do fabricante. Os dados obtidos foram analisados utilizando o programa Bioconductor e o algoritmo crlmm. Para os SNPs, as diferenças entre os grupos foram analisadas por meio da regressão logística múltipla. Para as CNVs, os dados obtidos foram analisados por meio do programa Partek®. Regiões de ganhos ou perdas significativas de DNA foram determinadas pelo algoritmo cbs. Os genes de interesse foram escolhidos por meio do programa DAVID. Nós observamos que a frequência de 6.609 SNPs foi distinta entre pacientes com CEC de BL e controles (P< 0,01). Cinquenta e dois SNPs (0,8%) estiveram localizados nas regiões codificantes do DNA, 51 (0,8%) estiveram nas regiões 3' e 5' não traduzidas, 3.461 estiveram em regiões regulatórias de transcrição e 3.045 em íntrons. Os SNPs considerados de interesse estiveram localizados nos genes relacionados ao ciclo celular (ERP29, MCC e PTCH1), à transcrição (IKBKAP e ZNF415) e à adesão celular (COL22A1, LEF1 e LY6K). Nós identificamos regiões do DNA que apresentaram duplicações em genes relacionados com a proliferação celular (ADAM3A, ADAM5P e DDT), apoptose (FAM90A), mecanismo de defesa (DEFB) e metabolismo de carcinógenos (GSTs). Nós também observamos genes deletados relacionados à apoptose (BLC2) e aos receptores do olfato (ORs). Nossos resultados sugerem que SNPs e CNVs em genes relacionados com a origem e a progressão de tumores podem predispor indivíduos ao CEC de BL. No entanto, esses resultados devem ser validados por genotipagens de número maior de indivíduos e por análises funcionais de proteínas codificadas por alelos distintos de genes polimórficos / Abstract: Inherited genetic alterations, such as single nucleotide polymorphisms (SNPs) and copy number variations (CNVs), were described in association with base of tongue (BT) squamous cell carcinoma (SCC) risk in only few reports. BTSCC are tumours with high morbidity and mortality rates; however, the association of SNPs and CNVs and BTSCC risk is still not clarified and, therefore, this was the aim of the present study. DNA was extracted of the peripheral blood samples of 49 BTSCC patients and 49 controls. Each sample was genotyped using DNA high-resolution microarrays containing 500.568 SNPs and 420.000 CNVs (Affymetrix®). Further sample processing, including digestion, adaptor ligation, amplification, fragmentation, labelling, hybridization, washing and scanning was assayed according to the standard protocol. Genotype data were acquired by genotyping calling of samples using the crlmm algorithm provided by Bioconductor software, as per the recommended guidelines. For SNPs, the differences between groups were analysed by the logistic regression model. For CNVs, the patients' and controls' data files were imported into the Partek® Genomic Suite. Common aberration analysis was performed on all samples to identify genomic intervals that had statistically significant aberrations. Significantly different regions were determined using the segmentation algorithm. For SNPs, we observed 6.609 SNPs with distinct frequencies between BTSCC patients and controls (P< 0.01). Fifty two SNPs (0.8%) were located in coding sequence of amino acids, 51 (0.8%) in 3' and 5' untranslated regions, 3.461 (52.4%) in up or downstream regions and 3.045 (46.0%) in introns. The SNPs were clustered to their main function, evidencing those localized in genes related to cell cycle and apoptosis (ERP29, MCC and PTCH1), transcriptional process (IKBKAP and ZNF415) and cell adhesion and metastasis (COL22A1, LEF1 and LY6K). We also identified a consistent number of altered regions including duplicated genes, such as involved in cell proliferation and angiogenesis (ADAM3A, ADAM5P and DDT), apoptosis (FAM90A), defensins proteins (DEFB) and metabolism of carcinogens (GSTs); and deleted genes, such as in olfactory receptors (ORs) and apoptosis (BCL2). Our preliminary results suggest that SNPs and CNVs in genes involved in tumour origin and progression may predispose individuals to BTSCC. However, these results should be confirmed by functional studies of coded proteins and validated by genotyping in larger epidemiological studies / Doutorado / Biologia Estrutural, Celular, Molecular e do Desenvolvimento / Doutor em Fisiopatologia Medica

The study of how social organizations work, change and develop is central to sociology and to our understanding of the social world and its transformations. At the same time, the underlying principles of organizational dynamics are extremely difficult to investigate. This is partly due to the difficulties of tracking organizations, individuals and their interactions over relatively long periods of time. But it is also due to limitations in the kinds of quantitative methods used to tackle these questions, which are for the most part based on regression analysis. This thesis seeks to improve our understanding of social organizing by using models to explore and describe the logics of the structures and mechanisms underlying organizational change. Particular emphasis is given to the modeling process, the use of new concepts and analogies, and the application of interdisciplinary methods to get new insights into classical sociological questions. The thesis consists of an introductory part and five studies (I-V). Using Swedish longitudinal data on employment in the Stockholm Region, the studies tackle different dimensions of organizational dynamics, from organizational structures and growth processes to labor mobility and employment trajectories. The introductory chapters contextualize the studies by providing an overview of theories, concepts and quantitative methods that are relevant for the modeling of organizational dynamics.  The five studies look into various aspects of organizational dynamics with the help of complementary data representations and non-traditional quantitative methods. Study I analyzes organizational growth statistics for different sectors and industries. The typically observed heavy-tailed statistical patterns for the size and growth rate distributions are broken down into a superposition of interorganizational movements. Study II models interorganizational movements as a labor flow network. Organizations tend to be more tightly linked if they belong to the same ownership sector. Additionally, public organizations have a more stable connection structure. Study III uses a similarity-based method called homogeneity analysis to map out the social space of large organizations in the Stockholm Region. A social distance is then derived within this space, and we find that the interorganizational movements analyzed in Studies I and II take place more often between organizations that are closer in social space and in the same network community. Study IV presents an approach to organizational dynamics based on sequences of employment states. Evidence for a positive feedback mechanism is found for large and highly sequence-diverse public organizations. Finally, Study V features an agent-based model where we simulate a social influence mechanism for organizational membership dynamics. We introduce a parameter analogous to a physical temperature to model contextual influence, and the familiar growth distributions are recovered as an intermediate case between extreme parameter values. The thesis as a whole provides suggestions for a more process-oriented modeling approach to social organizing that gives a more prominent role to the logics of organizational change. Finally, the series of methodological tools discussed can be useful for the analysis of many other social processes and more broadly for the development of quantitative sociological methods. / <p>At the time of the doctoral defense, the following papers were unpublished and had a status as follows: Paper 1: Manuscript. Paper 2: Manuscript. Paper 3: Manuscript. Paper 4: Manuscript.</p><p> </p>

Position weight matrices (PWMs) have become a tool of choice for the identification of transcription factor binding sites in DNA sequences. PWMs are compiled from experimentally verified and aligned binding sequences. PWMs are then used to computationally discover novel putative binding sites for a given protein. DNA-binding proteins often show degeneracy in their binding requirement, the overall binding specificity of many proteins is unknown and remains an active area of research. Although PWMs are more reliable predictors than consensus string matching, they generally result in a high number of false positive hits. A previous study introduced a novel method to PWM training based on the known motifs to sample additional putative binding sites from a proximal promoter area. The core idea was further developed, implemented and tested in this thesis with a large scale application. Improved mono- and dinucleotide PWMs were computed for Drosophila melanogaster. The Matthews correlation coefficient was used as an optimization criterion in the PWM refinement algorithm. New PWMs keep an account of non-uniform background nucleotide distributions on the promoters and consider a larger number of new binding sites during the refinement steps. The optimization included the PWM motif length, the position on the promoter, the threshold value and the binding site location. The obtained predictions were compared for mono- and dinucleotide PWM versions with initial matrices and with conventional tools. The optimized PWMs predicted new binding sites with better accuracy than conventional PWMs.

Pour les patients souffrants d’une maladie pulmonaire obstructive chronique (MPOC), il existe de nombreuses façons de cheminer dans le système de santé québécois. Toutefois, ces trajectoires de soins (TS) ne sont pas bien décrites dans la littérature et leur impact sur la mortalité demeure inconnu. Les objectifs du projet sont d’identifier et de caractériser les TS des patients MPOC, puis d’en évaluer l’association avec la mortalité. Pour ce faire, nous avons créé une cohorte rétrospective de 3352 Québécois, atteints d’une MPOC, ayant répondu à l’Enquête sur la santé des collectivités canadiennes (ESCC) entre 2007 et 2013. Trois banques de données administratives du Québec (RAMQ, MED-ÉCHO et RED/D) ont été jumelées aux données de l’ESCC pour mesurer les TS sur 2 ans, et ce, à l’aide de l’analyse de séquences. Les caractéristiques des patients ont été comparées entre les TS grâce au test du khi-2 et au V-test. Nous avons réalisé des modèles de Cox, bruts et ajustés pour les facteurs confondants, dans le but d’évaluer l’association entre les TS et la mortalité sur un maximum de 5 ans. Suite à l’analyse de séquences, 6 TS ont été identifiées : 1) faible utilisation des services de santé: toutes causes (44,7%); 2) utilisation moyenne des services de santé: toutes causes (27,7%); 3) grande utilisation des services de santé surtout chez l’omnipraticien: causes respiratoires (4,4%); 4) grande utilisation des services de santé surtout chez l’omnipraticien: autres causes que respiratoires (6,5%); 5) grande utilisation des services de santé surtout chez le spécialiste: autres causes que respiratoires (6,3%); 6) grande utilisation des services de santé surtout en soins aigus: toutes causes (10,4%). Les caractéristiques qui ont permis de distinguer l’appartenance des patients à une TS sont l’âge, le statut tabagique et matrimonial, le niveau d’éducation, l’indice de comorbidité et la santé perçue. Les patients des TS 2, 4 et 6 étaient significativement plus à risque de décès que les patients de la TS 1. Ainsi, une meilleure compréhension des TS et de leur association avec la mortalité pourrait aider les décideurs et les cliniciens à mieux identifier les patients à risque et allouer judicieusement les services et les ressources afin d’optimiser la gestion de la MPOC. / Patients with chronic obstructive pulmonary disease (COPD) navigate in the healthcare system in different ways. However, those trajectories of care (ToCs) are not well described and their impact on mortality is unknown. Therefore, our research project’s objectives were to identify and characterize the ToCs of COPD patients and to evaluate the association between those ToCs and mortality. To do so, we conducted a retrospective cohort study with 3352 COPD patients living in Quebec, who answered the Canadian Community Health Survey between 2007 and 2013. Three Quebec health administrative databases (RAMQ, MED-ÉCHO and RED/D) were linked to the survey data to measure ToCs over a 2-year period, using sequence analysis. Patients’ characteristics were compared between ToCs with chi-square and V tests. At last, we evaluated the association between ToCs and death over a maximum of 5 years, using Cox models adjusted for potential confounders. Following sequence analysis, 6 ToCs were identified: 1) low use of healthcare services: all causes (44.7%); 2) medium use of healthcare services: all causes (27.7%); 3) high use of healthcare services mainly with general practitioners (GPs): respiratory causes (4.4%); 4) high use of healthcare services mainly with GPs: other causes than respiratory (6.5%); 5) high use of healthcare services mainly with specialists: other causes than respiratory (6.3%) and 6) high use of healthcare services mainly in acute care: all causes (10.4%). Patients’ characteristics that best-described the trajectories’ membership were age, smoking status, marital status, level of education, comorbidity index and, patients’ perceived health. Patients belonging to trajectories 2, 4 and 6 were significantly more at risk of death than patients belonging to the first trajectory. Thus, a better understanding of the ToCs in Quebec and their association with mortality could help policymakers and clinicians for the allocation of care and health resources, thereby promoting least fatal ToCs.

<p>Advances in Artificial Intelligence (AI), or more precisely on Neural Networks (NNs), and fast processing technologies (e.g. Graphic Processing Units or GPUs) in recent years have positioned NNs as one of the main machine learning algorithms used to solved a diversity of problems in both academia and the industry. While they have been proved to be effective in solving many tasks, the lack of security guarantees and understanding of their internal processing disrupts their wide adoption in general and cybersecurity-related applications. In this dissertation, we present the findings of a comprehensive study aimed to enable the absorption of state-of-the-art NN algorithms in the development of enterprise solutions. Specifically, this dissertation focuses on (1) the development of defensive mechanisms to protect NNs against adversarial attacks and (2) application of NN models for anomaly detection in enterprise networks.</p><p>In this state of affairs, this work makes the following contributions. First, we performed a thorough study of the different adversarial attacks against NNs. We concentrate on the attacks referred to as trojan attacks and introduce a novel model hardening method that removes any trojan (i.e. misbehavior) inserted to the NN models at training time. We carefully evaluate our method and establish the correct metrics to test the efficiency of defensive methods against these types of attacks: (1) accuracy with benign data, (2) attack success rate, and (3) accuracy with adversarial data. Prior work evaluates their solutions using the first two metrics only, which do not suffice to guarantee robustness against untargeted attacks. Our method is compared with the state-of-the-art. The obtained results show our method outperforms it. Second, we proposed a novel approach to detect anomalies using LSTM-based models. Our method analyzes at runtime the event sequences generated by the Endpoint Detection and Response (EDR) system of a renowned security company running and efficiently detects uncommon patterns. The new detecting method is compared with the EDR system. The results show that our method achieves a higher detection rate. Finally, we present a Moving Target Defense technique that smartly reacts upon the detection of anomalies so as to also mitigate the detected attacks. The technique efficiently replaces the entire stack of virtual nodes, making ongoing attacks in the system ineffective.</p><p> </p>