Translation of technical manuals is expensive, especially when a larger company needs to publish manuals for their whole product range in over 20 different languages. When a text segment (i.e. a phrase, sentence or paragraph) is manually translated, we would like to reuse these translated segments in future translation tasks. A translated segment is stored with its corresponding source language, often called a language pair in a Translation Memory System. A language pair in a Translation Memory represents a Translation Entry also known as a Translation Unit. During a translation, when a text segment in a source document matches a segment in the Translation Memory, available target languages in the Translation Unit will not require a human translation. The previously translated segment can be inserted into the target document. Such functionality is provided in the single source publishing software, Skribenta developed by Excosoft. Skribenta requires text segments in source documents to find an exact or a full match in the Translation Memory, in order to apply a translation to a target language. A full match can only be achieved if a source segment is stored in a standardized form, which requires manual tagging of entities, and often reoccurring words such as model names and product numbers. This thesis investigates different ways to improve and optimize a Translation Memory System. One way was to aid users with the work of manual tagging of entities, by developing Heuristic algorithms to approach the problem of Named Entity Recognition (NER). The evaluation results from the developed Heuristic algorithms were compared with the result from an off the shelf NER tool developed by Stanford. The results shows that the developed Heuristic algorithms is able to achieve a higher F-Measure compare to the Stanford NER, and may be a great initial step to aid Excosofts’ users to improve their Translation Memories. / Översättning av tekniska manualer är väldigt kostsamt, speciellt när större organisationer behöver publicera produktmanualer för hela deras utbud till över 20 olika språk. När en text (t.ex. en fras, mening, paragraf) har blivit översatt så vill vi kunna återanvända den översatta texten i framtida översättningsprojekt och dokument. De översatta texterna lagras i ett översättningsminne (Translation Memory). Varje text lagras i sitt källspråk tillsammans med dess översättning på ett annat språk, så kallat målspråk. Dessa utgör då ett språkpar i ett översättningsminnessystem (Translation Memory System). Ett språkpar som lagras i ett översättningsminne utgör en Translation Entry även kallat Translation Unit. Om man hittar en matchning när man söker på källspråket efter en given textsträng i översättningsminnet, får man upp översättningar på alla möjliga målspråk för den givna textsträngen. Dessa kan i sin tur sättas in i måldokumentet. En sådan funktionalitet erbjuds i publicerings programvaran Skribenta, som har utvecklats av Excosoft. För att utföra en översättning till ett målspråk kräver Skribenta att text i källspråket hittar en exakt matchning eller en s.k. full match i översättningsminnet. En full match kan bara uppnås om en text finns lagrad i standardform. Detta kräver manuell taggning av entiteter och ofta förekommande ord som modellnamn och produktnummer. I denna uppsats undersöker jag hur man effektivt implementerar en optimering i ett översättningsminnessystem, bland annat genom att underlätta den manuella taggningen av entitier. Detta har gjorts genom olika Heuristiker som angriper problemet med Named Entity Recognition (NER). Resultat från de utvecklade Heuristikerna har jämförts med resultatet från det NER-verktyg som har utvecklats av Stanford. Resultaten visar att de Heuristiker som jag utvecklat uppnår ett högre F-Measure jämfört med Stanford NER och kan därför vara ett bra inledande steg för att hjälpa Excosofts användare att förbättra deras översättningsminnen.

Résumé : Le nucléole est considéré comme étant une « usine » à produire des ribosomes. Cette production est la fonction la plus énergivore de la cellule. Elle met en jeu les trois ARN polymérases et représente 80% de l’activité de transcription au sein d’une cellule. Les trois quarts de cette activité de transcription correspondent à la synthèse des ARNr par l’ARN polymérase I (ARNPI). Ainsi mieux comprendre les mécanismes cellulaires se déroulant à l’intérieur de ce compartiment permettra le développement de nouveaux traitements contre le cancer. La synthèse d’ARNr par l’ARNPI est régulée à trois niveaux : l’initiation de la transcription, l’élongation et le nombre de gènes de l’ARNr en transcription. La plupart des travaux qui se sont intéressés à ces niveaux de régulation ont été réalisés avec des cellules en phase exponentielle de croissance. Au cours de mes travaux, je me suis attardé sur la régulation de la transcription par l’ARNPI au cours de la phase G1 du cycle cellulaire et au début de la phase S. Ainsi mes résultats ont montré que si la chromatine des gènes de l’ARNr est essentiellement dépourvue de nucléosomes, la régulation de l’ARNPI diffère dans des cellules en G1 et au début de la phase S. J’ai pu de ce fait observer qu’en G1, la transcription de l’ARNPI se concentre sur un nombre réduit de gènes en transcription. Dans des cellules arrêtées au début de la phase S avec de l’hydroxyurée, la transcription de l’ARNPI est perturbée par un défaut de maturation de l’ARNR. Fort de ces résultats sur la nature des gènes ribosomaux en phase G1, je me suis attardé à la réparation de ces gènes lors de cette phase. Alors que dans des cellules en phase exponentielle de croissance irradiées avec des UVC, la chromatine des gènes de l’ARNr se ferme ; je n’ai pas observé la formation de nucléosomes suite à l’irradiation de cellules synchronisée en G1. Mes résultats montrent également que la réparation est plus efficace. Parallèlement, j’ai exploré l’assemblage du complexe de réparation par excision de nucléotides. Toutefois, les résultats obtenus sont peu concluants. / Abstract : The nucleolus is thought to be a “factory” involve in the production of ribosomes. This production is the most energetically consuming process in the cell. The three RNA polymerases are involved and this represents 80% of the total transcription activity of the cell. Three quarters of this transcriptional activity correspond to the synthesis of rRNA by the RNA polymerase I (RNAPI). So a better understanding of the cellular mechanisms taking place in this compartment may help for the development of new drugs against cancer. The synthesis of rRNA by RNAPI is regulated at three levels: initiation of transcription, elongation and the number of rRNA genes in transcription. Most of the works that characterized those levels of regulation were done in exponentially growing cells. During my work, I studied the regulation of RNAPI during the G1 phase of the cell cycle and during the early S phase. So my results have shown that if the chromatin of the rRNA genes mostly depleted of nucleosomes, the regulation of the RNAPI differs in cells in G1 and early S phase. I could observe that in G1, RNAPI transcription concentrates on a reduced number of transcribed rRNA genes. In cells arrested in early S phase with hydroxyurea, RNAPI transcription is disrupted by a defect in rRNA processing. With this results on the nature of the ribosomal genes in G1, I started the analysis of the DNA repair of those genes during this phase of the cell cycle. In UVC irradiated exponentially growing cells, the rRNA genes are closing. But in cells synchronized in G1, I could not observe the deposition of nucleosomes after UVC irradiation. Moreover, my results show an increase repair of the locus. In parallel, I have explored the assembly of the complex of nucleotide excision repair. However, the results were not conclusive.

Mon sujet de thèse repose sur une observation initiale réalisée sur le panel de 60 lignées cellulaires tumorales humaines du National Cancer Institute (NCI) et concerne des polymorphismes de gènes impliqués dans la réponse aux agents anticancéreux, les gènes ERCC2 et ERCC5 ( à l'origine des protéines XPD et XPG) qui interviennent dans la réparation de l'ADN par le mécanisme du Nucléotide Excicion Repair (NER). Nous avons identifié une association entres ces polymorphismes et la cytotoxicité in vitro d'agents anticancéreux, en particulier celle des taxanes. A partir de ce travail initial, nos recherches se sont établies dans deux directions : (1) une approche clinique visant à chercher un lien significatif entre ces deux polymorphismes et la sensibilité aux taxanes chez des patients traités par chimiothérapie; (2) une approche biochimique visant à comprendre les mécanismes moléculaires expliquant l'association entre ces deux polymorphismes et la sensibilité cellulaire aux taxanes. Ce travail a permis de montrer que le niveau d'expression de ERCC2 modifié par le polymorphisme du codon 751, ainsi que la variation protéique entrainée par le polymorphisme du codon 1104 de ERCC5, étaient impliqués dans la régulation du cycle cellulaire par modification de l'activité CDK1. La réponse au paclitaxel se trouve ainsi modifiée du fait de ce mécanisme. A défaut d'être définitifs, nos résultats de l'étude clinique sont encourageants et montrent une tendance à une probabilité de réponse plus élevée chez les patientes de génotype ERCC2 homozygote variant et chez les patientes de génotype ERCC5 homozygote sauvage ou hétéroygote lorsque la chimiothérapie contient un taxane, le docetaxel.

No description available.

Diseases take a central role in biomedical research; many studies aim to enable access to disease information, by designing named entity recognition models to make use of the available information. Disease recognition is a problem that has been tackled by various approaches of which the most famous are the lexical and supervised approaches. However, the aforementioned approaches have many drawbacks as their performance is affected by the amount of human-annotated data set available. Moreover, lexicalapproachescannotdistinguishbetweenrealmentionsofdiseasesand mentionsofotherentitiesthatsharethesamenameoracronym. Thechallengeofthis project is to find a model that can combine the strengths of the lexical approaches and supervised approaches, to design a named entity recognizer. We demonstrate that our model can accurately identify disease name mentions in text, by using word embedding to capture context information of each mention, which enables the model todistinguishifitisarealdiseasementionornot. Weevaluateourmodelusingagold standard data set which showed high precision of 84% and accuracy of 96%. Finally, we compare the performance of our model to different statistical name entity recognition models, and the results show that our model outperforms the unsupervised lexical approaches.

Il est important d'évaluer régulièrement les produits de l'innovation technologique afin d'estimer le niveau de maturité atteint par les technologies et d'étudier les cadres applicatifs dans lesquels elles pourront être exploitées. Le traitement automatique des langues (TAL) relève à la fois de la recherche et de l'innovation technologique et a pour but la modélisation et le développement d'outils permettant de traiter automatiquement le langage naturel. Pendant longtemps, les différentes briques technologiques issues du TAL étaient développées séparément. Par conséquent, les méthodes d'évaluation existantes sont dans la plupart modulaires et ne permettent d'évaluer qu'un seul module à la fois, alors qu'aujourd'hui nombreuses applications nécessitent de combiner plusieurs modules de TAL pour résoudre des tâches complexes. Le nouveau défi en terme d'évaluation est alors de pouvoir évaluer les différents modules (ou briques) tout en prenant en compte le contexte applicatif.Notre travail porte sur l'évaluation des systèmes de reconnaissance automatique de la parole (RAP) en contexte applicatif, en particulier, celui de la reconnaissance d'entités nommées (REN).En première partie, nous abordons la problématique de l'évaluation des systèmes de RAP en contexte applicatif à travers une étude de l'état de l'art. Nous y décrivons les tâche de RAP et de REN proposées dans les campagnes d'évaluation ainsi que les protocoles mis en place pour leurs évaluation. Nous y discutons également les limites des approches d'évaluations modulaires et nous y exposons les mesures alternatives proposées dans la littératures. En deuxième partie, nous décrivons la tâche de détection, classification et décomposition d'entités nommées étudiée et nous proposons une nouvelle métriques ETER (Entity Tree Error Rate) permettant de prendre en compte les spécificité de cette tâche et le contexte applicatif lors de l'évaluation. ETER permet également de supprimer les biais observés avec les métriques existantes. En troisième partie, nous définissons une nouvelle mesure ATENE (Automatic Transcriptions Evaluation for Named Entities) qui permet d'évaluer la qualité des systèmes de RAP et l'impact de leurs erreurs pour des systèmes de REN appliqués en aval. ATENE consiste à comparer les probabilités de présence d'entités sur les transcriptions de référence et d'hypothèse plutôt qu'une comparaison directe des graphèmes. Elle est composée de deux mesures élémentaires. Une première permettant l'évaluation de risque d'erreur d'omission et de substitution d'entités et une seconde permettant d'évaluer le risque d'erreur d'insertion d'entités causé par les erreurs de RAP.Nos expériences de validation montrent que les mesures données par ATENE corrèlent mieux que les autres mesures de l'état de l'art avec les performances des systèmes de REN. / It is important to regularly assess the technological innovation products in order to estimate the level of maturity reached by the technology and study the applications frameworks in which they can be used. Natural language processing (NLP) aims at developing modules and applications that automatically process the human language. That makes the field relevant to beth research and technological innovation. For years, the different technological modules from the NLP were developed separately. Therefore, the existing evaluation methods are in most modular. They allow to evaluate only one module at a time, while today, many applications need to combine several NLP modules to solve complex tasks. The new challenge in terms of evaluation is then to evaluate the different modules while taking into account the applicative context.Our work addresses the evaluation of Automatic Speech Recognition (ASR) systems according to the applicative context. We will focus on the case of Named Entities Recognition (NER) from spoken documents transcriped automatically. In the first part, we address the issue of evaluating ASR systems according to the application context through a study of the state of the art. We describes the tasks of ASR and NER proposed during several evalution campaigns and we discuss the protocols established for their evaluation. We also point the limitations of modular evaluation approaches and we expose the alternatives measures proposed in the literature. In the second part we describe the studied task of named entities detection, classification and decomposition and we propose a new metric ETER (Entity Tree Error Rate) which allows to take into account the specificity of the task and the applicative context during the evaluation. ETER also eliminates the biases observed with the existing metrics. In the third part, we define a new measure ATENE (Automatic Transcriptions Evaluation for Named Entities) that evaluates the quality of ASR systems and the impact of their errors for REN systems applied downstream. Rather than directly comparing reference and hypothesis transcriptions, ATENE measure how harder it becames to identify entities given the differences between hypothesis and reference by comparing an estimated likelihood of presence of entities. It is composed of two elementary measurements. The first aims to assess the risk of entities deletions and substitutions and the second aims to assess the risk of entities insertions caused by ASR errors.Our validation experiments show that the measurements given by ATENE correlate better than other measures from the state of the art with the performance of REN systems.

The amount of data and analysis being published and archived in the biomedical research community is more than can feasibly be sifted through manually, which limits the information an individual or small group can synthesize and integrate into their own research. This presents an opportunity for using automated methods, including Natural Language Processing (NLP), to extract important information from text on various topics. Named Entity Recognition (NER), is one way to automate knowledge extraction of raw text. NER is defined as the task of identifying named entities from text using labels such as people, dates, locations, diseases, and proteins. There are several NLP tools that are designed for entity recognition, but rely on large established corpus for training data. Biomedical research has the potential to guide diagnostic and therapeutic decisions, yet the overwhelming density of publications acts as a barrier to getting these results into a clinical setting. An exceptional example of this is the field of breast cancer biology where over 2 million people are diagnosed worldwide every year and billions of dollars are spent on research. Breast cancer biology literature and research relies on a highly specific domain with unique language and vocabulary, and therefore requires specialized NLP tools which can generate biologically meaningful results. This thesis presents a novel annotation tool, that is optimized for quickly creating training data for spaCy pipelines as well as exploring the viability of said data for analyzing papers with automated processing. Custom pipelines trained on these annotations are shown to be able to recognize custom entities at levels comparable to large corpus based recognition.

A quimiorresistência de tumores constitui um dos maiores obstáculos que levam comumente ao fracasso da terapia. Os mecanismos relevantes que contribuem para a resistência celular incluem: bombas de efluxo; alterações na interação entre a droga e o seu alvo e mudanças nas respostas celulares, em particular uma habilidade aumentada de reparar os danos induzidos no DNA e defeitos nas vias apoptóticas. A capacidade de reparar os danos no DNA e a evasão da apoptose são de grande importância, uma vez que a maioria dos quimioterápicos tem sua ação baseada na indução de citotoxicidade pela capacidade de gerar lesões no DNA. Desta forma, uma importante estratégia para melhorar a quimioterapia é o desenvolvimento de abordagens mais seletivas e mecanismos que contornem a resistência tumoral. Neste trabalho, através de um estudo sobre os genes e suas respectivas vias envolvidas no reparo, capacidade de sobrevivência e sinalização de danos induzidos pela nimustina (ACNU) - um agente cloroetilante comumente utilizado em tratamentos quimioterápicos de tumores sólidos - identificamos genes potencialmente alvos para uma terapia adjuvante. Demonstramos que células de glioma p53mt tem menor capacidade de reparo de ICLs induzidos por esta droga do que células p53wt. Também, que a via de NHEJ (\'\'Non Homologous End Joining\'\') não é uma via preferencial de reparo dessas lesões, mas que a via de NER (\'\'Nucleotide Excision Repair\'\') (ou especificamente os produtos gênicos XPA, XPC e XPF) é bastante importante. Curiosamente, na ausência de XPA, NHEJ assume uma participação no reparo dessas lesões, provavelmente devido a um aumento no número de DSBs e saturação das outras vias de reparo. Da mesma forma, verificamos que a DNA polimerase POLH (XPV), envolvida em TLS (\'\'Translesion Synthesis\'\'), também participa na tolerância dessas lesões. Neste contexto, encontramos evidências de que a polimerase TLS (especificamente POLH e POLK) apresentam-se superexpressas em amostras de gliomas, podendo desta forma concorrerem tanto para a tumorigênese quanto para a resistência observada nestes tipos tumorais. Por fim, realizamos o silenciamento gênico através da teconologia de RNAi, que reprimem os genes pela eliminação do transcrito mRNA correspondente, prevenindo a síntese protéica. Os genes-alvo escolhidos para o silenciamento foram, desta forma, XPC, XPF, POLH e POLK. O silenciamento gênico de XPC, XPF e POLH demonstraram-se capazes de sensibilizar significativamente células de glioma, permitindo-nos sugerir estas proteínas como elementos importantes na quimioresistência de gliomas ao ACNU e colocando a inibição dessas moléculas como uma estratégia importante na sensibilização de gliomas ao ACNU e potencialmente a outros agentes quimioterápicos com o mesmo mecanismo de ação. / The chemoresistance of tumors is one of the most important obstacles that commonly lead to the failure of therapy. The main mechanisms that contribute to cellular resistance include efflux pumps; changes in the interaction between the drug and its target and changes in cellular responses, in particular an increased ability to repair induced DNA damages and defects in apoptotic pathways. The ability to repair DNA damage and evasion of apoptosis are of great importance, since most chemotherapy has its action based on the induction of cytotoxicity by the ability to generate DNA lesions. Thus, an important strategy for improving chemotherapy is the development of more selective mechanisms that circumvent tumor resistance approaches. In this work, through a study of genes and pathways involved in the repair, survival and damage signaling induced by nimustine (ACNU) - a cloroethylating agent commonly used in treatments of solid tumors - we aimed to identify target genes for a potentially adjuvant therapy. We demonstrated that glioma cells p53mt have less ability to repair ICLs induced by this drug then p53wt cells. Also, that the NHEJ (\'\'Non Homologous End Joining\'\') pathway is not the main route of repair of these lesions, but that the NER (\'\'Nucleotide Excision Repair\'\') pathway (or specifically the gene products XPA, XPC and XPF) is very important. Interestingly, in the absence of XPA, NHEJ takes place in the repair of those lesions, probably due to an increase in the number of DSBs and saturation of other repair pathways. Likewise, we found that DNA polimerase involved in TLS (\'\'Translesion Synthesis\'\') POLH (XPV) also participates in tolerance of such lesions. We also found evidence that TLS polimerases (specifically POLH and POLK) are overexpressed in gliomas samples and could play a role in the tumorigenesis and in the resistance observed in these tumor types. Finally, we performed gene silencing through RNAi teconology, which repress genes by eliminating the corresponding mRNA transcript, preventing protein synthesis. The target genes selected for silencing were XPC, XPF, POLH and POLK. The knockdown of XPC, XPF and POLH proved to significantly sensitize glioma cells, suggesting these proteins as important elements in the chemoresistance of gliomas and highlighting the inhibition of these molecules as an important strategy in the sensitization of gliomas to ACNU and probably to other chemotherapeutic agents with the same mechanisms of action.

La voie de réparation NER implique une cascade de complexes protéiques dont le senseur des dommages de l’ADN (XPC/HR23B). Des mutations dans les gènes de la NER (TTD-A, XPA-G, XPV, CSA et CSB), sont associées à des maladies génétiques humaines dont le Xeroderma Pigmentosum (XP), la Trichothiodystrophie (TTD) et le syndrome de Cockayne (CS). L’ensemble des symptômes des patients ne peut être expliqué seulement par un défaut de la réparation de l’ADN. Or depuis quelques années, il a été prouvé que les facteurs de la NER sont aussi impliqués lors de la transcription. Dans le cadre de ma thèse, je me suis particulièrement intéressé à la protéines XPC en déterminant son rôle transcriptionnel à l’échelle génomique afin de mieux comprendre les conséquences de sa dérégulation dans un contexte pathologique. En ce sens, mon second objectif a été de caractériser au niveau moléculaire l’étiologie de nouveaux patients XP en analysant de manière combinée les évènements moléculaires de la NER et la transcription associés à XPC. Nos différentes approches expérimentales ont permis d’identifier au niveau génomique un ensemble de gènes sont les promoteurs sont régulés aussi bien positivement que négativement par XPC dans un contexte RAR dépendant. De plus, nous montrons que XPC interagit avec KAT2A contenu dans le complexe ATAC, ainsi que qu’avec le facteur de transcription E2F1, le facteur de remodelage de la chromatine BRD2 et le variant d’histone H2A.Z. Via KAT2A, ce complexe va acétyler non seulement H2A.Z mais également H3K9 au niveau des promoteurs ciblés par E2F1. / NER involves a cascade of protein complexes including the DNA damage sensor (XPC/HR23B). Mutations in NER genes (TTD-A, XPA-G, XPV, CSA and CSB) are associated with human genetic diseases including Xeroderma pigmentosum (XP), Trichothiodystrophy (TTD) and Cockayne Syndrome (CS). All the symptoms can only be explained by a defect of the DNA repair. However all the symptoms can only be explained by a defect of the DNA repair. However, it has been proven that NER factors are also involved in transcription. As the genomic scale to better understand the consequences of its deregulation in a pathological context. In this sense, my second goal has been to characterize at the molecular level the etiology of new XP patients by analyzing in a combined way the molecular events of the NER and the transcription associated with XPC. Our different experimental approaches have made it possible to identify at genomic level a set of gene whose promoters are regulated both positively and negatively by XPC in a dependent RAR context. In addition, we show that XPC interacts with KAT2A contained in the ATAC complex, as well as with the transcription factor E2F1, the chromatin remodeling factor BRD2, and the histone variant H2A.Z. Via KAT2A, this complex will acetylate not only H2A.Z but also H3K9 at promoters targeted by E2F1.

A quimiorresistência de tumores constitui um dos maiores obstáculos que levam comumente ao fracasso da terapia. Os mecanismos relevantes que contribuem para a resistência celular incluem: bombas de efluxo; alterações na interação entre a droga e o seu alvo e mudanças nas respostas celulares, em particular uma habilidade aumentada de reparar os danos induzidos no DNA e defeitos nas vias apoptóticas. A capacidade de reparar os danos no DNA e a evasão da apoptose são de grande importância, uma vez que a maioria dos quimioterápicos tem sua ação baseada na indução de citotoxicidade pela capacidade de gerar lesões no DNA. Desta forma, uma importante estratégia para melhorar a quimioterapia é o desenvolvimento de abordagens mais seletivas e mecanismos que contornem a resistência tumoral. Neste trabalho, através de um estudo sobre os genes e suas respectivas vias envolvidas no reparo, capacidade de sobrevivência e sinalização de danos induzidos pela nimustina (ACNU) - um agente cloroetilante comumente utilizado em tratamentos quimioterápicos de tumores sólidos - identificamos genes potencialmente alvos para uma terapia adjuvante. Demonstramos que células de glioma p53mt tem menor capacidade de reparo de ICLs induzidos por esta droga do que células p53wt. Também, que a via de NHEJ (\'\'Non Homologous End Joining\'\') não é uma via preferencial de reparo dessas lesões, mas que a via de NER (\'\'Nucleotide Excision Repair\'\') (ou especificamente os produtos gênicos XPA, XPC e XPF) é bastante importante. Curiosamente, na ausência de XPA, NHEJ assume uma participação no reparo dessas lesões, provavelmente devido a um aumento no número de DSBs e saturação das outras vias de reparo. Da mesma forma, verificamos que a DNA polimerase POLH (XPV), envolvida em TLS (\'\'Translesion Synthesis\'\'), também participa na tolerância dessas lesões. Neste contexto, encontramos evidências de que a polimerase TLS (especificamente POLH e POLK) apresentam-se superexpressas em amostras de gliomas, podendo desta forma concorrerem tanto para a tumorigênese quanto para a resistência observada nestes tipos tumorais. Por fim, realizamos o silenciamento gênico através da teconologia de RNAi, que reprimem os genes pela eliminação do transcrito mRNA correspondente, prevenindo a síntese protéica. Os genes-alvo escolhidos para o silenciamento foram, desta forma, XPC, XPF, POLH e POLK. O silenciamento gênico de XPC, XPF e POLH demonstraram-se capazes de sensibilizar significativamente células de glioma, permitindo-nos sugerir estas proteínas como elementos importantes na quimioresistência de gliomas ao ACNU e colocando a inibição dessas moléculas como uma estratégia importante na sensibilização de gliomas ao ACNU e potencialmente a outros agentes quimioterápicos com o mesmo mecanismo de ação. / The chemoresistance of tumors is one of the most important obstacles that commonly lead to the failure of therapy. The main mechanisms that contribute to cellular resistance include efflux pumps; changes in the interaction between the drug and its target and changes in cellular responses, in particular an increased ability to repair induced DNA damages and defects in apoptotic pathways. The ability to repair DNA damage and evasion of apoptosis are of great importance, since most chemotherapy has its action based on the induction of cytotoxicity by the ability to generate DNA lesions. Thus, an important strategy for improving chemotherapy is the development of more selective mechanisms that circumvent tumor resistance approaches. In this work, through a study of genes and pathways involved in the repair, survival and damage signaling induced by nimustine (ACNU) - a cloroethylating agent commonly used in treatments of solid tumors - we aimed to identify target genes for a potentially adjuvant therapy. We demonstrated that glioma cells p53mt have less ability to repair ICLs induced by this drug then p53wt cells. Also, that the NHEJ (\'\'Non Homologous End Joining\'\') pathway is not the main route of repair of these lesions, but that the NER (\'\'Nucleotide Excision Repair\'\') pathway (or specifically the gene products XPA, XPC and XPF) is very important. Interestingly, in the absence of XPA, NHEJ takes place in the repair of those lesions, probably due to an increase in the number of DSBs and saturation of other repair pathways. Likewise, we found that DNA polimerase involved in TLS (\'\'Translesion Synthesis\'\') POLH (XPV) also participates in tolerance of such lesions. We also found evidence that TLS polimerases (specifically POLH and POLK) are overexpressed in gliomas samples and could play a role in the tumorigenesis and in the resistance observed in these tumor types. Finally, we performed gene silencing through RNAi teconology, which repress genes by eliminating the corresponding mRNA transcript, preventing protein synthesis. The target genes selected for silencing were XPC, XPF, POLH and POLK. The knockdown of XPC, XPF and POLH proved to significantly sensitize glioma cells, suggesting these proteins as important elements in the chemoresistance of gliomas and highlighting the inhibition of these molecules as an important strategy in the sensitization of gliomas to ACNU and probably to other chemotherapeutic agents with the same mechanisms of action.