En plus des facteurs environnementaux et démographiques, les propriétés génétiques des populations sont devenues une préoccupation majeure pour préserver les populations en déclin de l'extinction. Afin d’acquérir des informations pertinentes pour la planification et la mise en œuvre des stratégies de conservation, les biologistes de la conservation ont réalisé le besoin d’avoir des connaissances en génétique des populations. Grace à l'acquisition de plus en plus rapide et de moins en moins chère d'une large gamme de marqueurs moléculaires, le recours a l’usage de l’outil moléculaire se répand de plus en plus. Ainsi, la génétique de la conservation se confirme comme une discipline à part entière qui est donc l’utilisation de la génétique dans la préservation des espèces comme entités dynamiques capables d'évoluer pour faire face aux changements environnementaux et afin de minimiser leur risque d'extinction. Par le biais de l’utilisation d’un large panel de marqueurs moléculaires (gènes mitochondriaux et nucléaires, microsatellites et SNPs), nous nous sommes intéresse à l’histoire évolutive à différentes échelles spatio-temporelles de la sous-espèce ouest méditerranéenne Testudo. hermanni hermanni (THH), qui présente une distribution insulaire et continentale très fragmentée. Le but de ce travail consiste à 1) comprendre les processus qui expliqueraient la distribution actuelle de la diversité génétique des populations et leur structure, 2) identifier l'origine des populations introduites (à Minorque et au Delta de l’Ebre), et 3) dater l’origine de la sous espèce THH. A l’échelle des populations, il s’agit d’identifier le nombre de groupes génétiques homogènes chez la tortue d’Hermann et le degré de différentiation génétique entre ces groupes afin de définir des unités de conservation évolutivement significatives (ESU) et des unités de gestion (MU). Enfin, nous nous sommes intéresses à l’étude des derniers noyaux de populations de THH dans le Var par des approches de génétique du paysage. Nos résultats ont révélé qu’une divergence par vicariance est à l’ origine de l’apparition de la sous-espèce T.h. hermanni. Ce scenario biogéographique s’expliquerait par les successions d’évènements glaciaires et interglaciaires qu’a connu le Pléistocène depuis plus de 2 MA provoquant un mouvement de retrait de l’espèce vers des zones refuges sur la frange côtière nord-méditerranéenne. Par ailleurs, le patron de différentiation mitochondriale Ile-continent observe et confirme par les microsatellites est très original par rapport à ce qui est connu chez d’autres espèces de reptiles partageant la même aire de distribution. Au vue de l’analyse phylogénétique confirmée par les microsatellites, on peut affirmer que la tortue d’Hermann n’est pas native sur Minorque et qu’elle a une double origine : la première, résultant d’une introduction à partir d’une seule source, probablement d’une population continentale génétiquement proche des Albères. La seconde d'origine insulaire, serait le résultat d’apports multiples, à partir de la Corse, de la Sardaigne ou de la Sicile. Enfin, l’isolement des populations de THH au sein de chaque région géographique reflète une structure génétique très forte. Par conséquent, six unités de gestion (MUs) sont proposées comme unités de conservation et de suivi sur le terrain. / In addition to environmental and demographic factors, the study of genetic properties of populations became inevitable issues in the conservation of declining populations. To acquire relevant information for conservation planning and implementing conservation strategies, conservationists have realized the need of population genetics tools. Moreover, this discipline has become more efficient with the development of a wide range of effective and relatively cheap methods for the characterization of a huge number of molecular markers. This led to define the conservation genetics as a separate discipline, which is the use of genetics in species preservation as dynamic entities evolving to cope with environmental changes and to minimize their extinction risk. Using a broad panel of molecular markers (mitochondrial and nuclear genes, microsatellites and SNPs), we interested in the evolutionary history at different spatial and temporal scales of the Mediterranean western subspecies Testudo hermanni hermanni (THH), which presents a very fragmented insular and continental distribution. The aim of this study is to 1) understand the processes that explain the current distribution of the structure and genetic diversity of populations, 2) identify the origin of introduced populations (Menorca and Ebro Delta) and 3) Dating the origin of the subspecies THH. At the population level, our study aimed to identify the number of homogeneous genetic groups of THH tortoise and the degree of genetic differentiation between these groups in order to identify evolutionarily significant units (ESU) and management units (MU). Finally, we were interested in the study of the last core populations of THH in the Var by landscape genetics approach. Our results revealed that a divergence by vicariance pattern explains the origin of the appearance of the subspecies THH. This biogeographic scenario is explained by the succession of glacial and interglacial events of the Pleistocene causing a withdrawal of the species toward refugia on the northern Mediterranean fringe. Moreover, the observed differentiation pattern (island vs continent) is very original compared to the reported diversity patterns of other reptiles sharing the same distribution range. According to our results, we may conclude that the Hermann’s tortoise is not native in Menorca and has a double origin: the first, is an introduction resulting from a unique source, probably from a continental lineage genetically close to Albera. The second, from an island origin, is the result of multiple contributions, from Corsica, Sardinia or Sicily. Lastly,the isolation of THH populations within each geographic region reflects a very strong genetic structure, therefor the six most relevant management units forconservation purposes are proposed on the basis that they represent a significant part of the evolutionary legacy of the species.

Introdução: O número de estudos relacionados à Síndrome Metabólica (SM) vem aumentando nos últimos anos, muitas vezes motivados pelo aumento do número de casos de sobrepeso/obesidade e diabetes Tipo II levando ao desenvolvimento de doenças cardiovasculares e, como consequência, infarto agudo do miocárdio e AVC, dentre outros desfechos desfavoráveis. A SM é uma doença multifatorial composta de cinco características, porém, para que um indivíduo seja diagnosticado com ela, possuir pelo menos três dessas características torna-se condição suficiente. Essas cinco características são: Obesidade visceral, caracterizada pelo aumento da circunferência da cintura, Glicemia de jejum elevada, Triglicérides aumentado, HDL-colesterol reduzido, Pressão Arterial aumentada. Objetivo: Estabelecer a rede de associações entre os fenótipos que compõem a Síndrome Metabólica através do aprendizado de estruturas de dependência, decompor a rede em componentes de correlação genética e ambiental e avaliar o efeito de ajustes por covariáveis e por variantes genéticas exclusivamente relacionadas à cada um dos fenótipos da rede. Material e Métodos: A amostra do estudo corresponderá a 79 famílias da cidade mineira de Baependi, composta por 1666 indivíduos. O aprendizado de estruturas de redes será feito por meio da Teoria de Grafos e Modelos de Equações Estruturais envolvendo o modelo linear misto poligênico para determinar as relações de dependência entre os fenótipos que compõem a Síndrome Metabólica / Introduction: The number of studies related to Metabolic Syndrome (MetS) has been increasing in the last years, encouraged by the increase on the overweight / obesity and Type II Diabetes cases, leading to the development of cardiovascular disease and, therefore, acute myocardial infarction and stroke, and others unfavorable outcomes. MetS is a multifactorial disease containing five characteristics, however, for an individual to be diagnosed with MetS, he/she may have at least three of them. These characteristics are: Truncal Obesity, characterized by increasing on the waist circumference, increasing on Fasting Blood Glucose, increasing on Triglycerides, decreasing on HDL cholesterol and increasing on Blood Pressure. Aims: Establish the best association network between MetS phenotypes through structured dependency learning between phenotypes considering genetic variants exclusively related to each phenotype. Materials and Methods: The study sample is composed of 79 families, 1666 individuals of a city in a rural area of Brazil, called Beapendi. Structured learning will use graph theory and Structural Equations Models to establish the dependency relations between MetS phenotypes

S'ha estudiat la variabilitat genètica de les poblacions del nord d'Àfrica a partir de l'anàlisi de diverses regions genòmiques per tal d'entendre les poblacions analitzades d'una banda, i comprendre la dinàmica del genoma per l'altra. Els resultats obtinguts ens han permès verificar diferents hipòtesis sobre la història de les poblacions d'aquesta regió com són l'efecte paral·lel i independent de l'onada de difusió del neolític des de l'Orient Mitjà al llarg d'ambdues ribes de la Mediterrànea; i l'efecte de l'arabització. S'ha pogut estimar també la contribució genètica masculina nord africana a la península ibèrica i detectat certa contribució genètica del pobles sub-saharians a les poblacions nordafricanes. Per altra banda, el tipatge de marcadors genètics que evolucionen a velocitats diferents al cromosoma Y ha permès mostrar que el background genètic predomina sobre el background poblacional en l'estructura de la variació genètica dels microsatèl·lits en la regió no recombinant del cromosoma Y humà. / The genetic variability of the North African populations has been studied through the analysis of different genomic regions in order to understand both the analysed populations and the dynamics of the genome. The obtained results allow us to verify different hypotheses about the population history of this region including the parallel and independent effect of the Neolithic wave of advance from the Middle East and along both Mediterranean coasts; and the effect of Arabization phenomena. We also tried to estimate the North African male genetic contribution to the Iberian peninsula and detected Sub-Saharian genetic influences to the North African peoples. Moreover, the typing of genetic markers with different evolutionary rates on the Y chromosome allowed us to demonstrate that variation in microsatellites is deeply structured by genetic background on the non-recombining region of the human Y chromosome.

Der plötzliche Kindstod (engl. Sudden Infant Death Syndrome-SIDS) ist die häufigste Todesursache bei Säuglingen innerhalb des ersten Lebensjahres. Die zugrundeliegenden pathophysiologischen Veränderungen sowie die genaue Todesursache sind bis dato ungeklärt. Viele Forschungsbereiche setzen sich intensiv mit der Klärung dieses „Phänomens“ auseinander. Ein Schwerpunkt liegt auf dem genetischen Gebiet und der Betrachtung verschiedener Polymorphismen. Ein Fokus wird hierbei auf die genetischen Polymorphismen der mitochondrialen DNA (mtDNA) gesetzt. In der vorliegenden Arbeit wurden daher drei Polymorphismen der mtDNA und mögliche Risikofaktoren im Bezug zu SIDS-Fällen untersucht. Die Folge eines mitochondrialen Polymorphismus kann beispielsweise die verminderte Genexpression der Untereinheiten der Atmungskette zur Folge haben. Daraus kann ein Defizit in der ATP (Adenosintriphosphat)-Produktion resultieren. Der physiologische Kreislauf einer menschlichen Zelle ist durch dieses Defizit nur eingeschränkt gewährleistet. Im Rahmen der Forschungsarbeit wurden die SNPs G3010A, T16519C und C7028T der mtDNA in Hinblick auf einen möglichen Zusammenhang mit dem SIDS untersucht. Schon 2003 untersuchten Divne et al. (2003) einen möglichen Zusammenhang der SNPs G3010A und C7028T im Zusammenhang mit SIDS, jedoch ohne signifikantes Ergebnis. Boles et al. (2010) konnten eine Assoziation zwischen den Polymorphismen G3010A und T16519C mit dem plötzlichen Kindstod herstellen. Da bislang jedoch keine ausführliche Publikation zu dieser Frage vorliegt, wurde mit der vorliegenden Arbeit die Rolle der Polymorphismen G3010A und T16519C in Bezug auf den plötzlichen Kindstod gemeinsam mit der (bei Europäern) häufigsten Variation C7028T untersucht. Die DNA von 176 SIDS-Fällen und einer Kontrollgruppe von 113 Erwachsenen wurde mittels Singleplex-PCR und RFLP-Analyse genotypisiert. Anhand der Genotypisierung konnten die SNPs quantifiziert und im Hinblick auf einen möglichen Unterschied zwischen SIDS-Fällen und der Kontrollgruppe untersucht werden. Bei Betrachtung der einzelnen SNPs G3010A, T16519C und C7028T lassen sich keine signifikanten Unterschiede zwischen den SIDS-Fällen und der Kontrollgruppe feststellen. Das gehäufte Vorliegen einer erhöhten Mutationsrate in einem Individuum bei SIDS-Fällen im Vergleich zur Kontrollgruppe, sowie die von Opdal et al. (1999) geäußerte Annahme, dass beim Vorliegen einer Mutation in der D-Loop-Region weitere Mutationen im kodierenden Bereich vorkommen, konnten durch diese Arbeit bestätigt werden.

No description available.

Made available in DSpace on 2016-06-02T20:20:30Z (GMT). No. of bitstreams: 1 3056.pdf: 1837497 bytes, checksum: cec4b528a2c534515e5a1880b1f0309d (MD5) Previous issue date: 2010-01-03 / Universidade Federal de Minas Gerais / The Canchim has been used in beef cattle as an alternative to production intensification. Canchim (5/8 Charolais + 3/8 Zebu) and MA (offspring of Charolais bulls and 1/2 Canchim + 1/2 Zebu cows) constitute a synthetic beef cattle breed which has a good growth potential and tropical adaptation but suboptimal fat deposition under pasture. Backfat thickness (BF), total fat amount and distribution of fat have a strong impact on carcass and meat quality in beef cattle. For this reason, research has been conducted to increase fat deposition in this breed, including the search for molecular markers to identify animals with high genetic potential for fat deposition. To incorporate molecular genetics into breeding programs in beef cattle, it is essential that the association between molecular markers and production traits is evaluated in the population in which they are to be used. There are reports of many candidate genes and chromosomal segments associated with variation in fat deposition in cattle. The objective of this study was to identify SNPs associated to backfat thickness in Canchim. To achieve this objective we searched for SNPs in development and differentiation enhancing factor 1 and insulin-like growth factor binding protein 3 genes, tested the association of SNPs in the insulin-like growth factor binding protein 3, peroxisome proliferative active receptor gamma coactivator 1A, proteasome 26S subunit ATPase 1, development and differentiation enhancing factor 1, corticotropin releasing hormone and fatty acid binding protein 4 genes with fat thickness in Canchim and MA beef cattle. We also analyzed the existence of genomic regions associated with backfat thickness in these populations using a 54 K chip in extreme phenotypes. From the associated regions we selected the ones of BTA14 to validate the association by analysis of haplotypes in the whole population. The SNPs analyzed in development and differentiation enhancing factor 1 and fatty acid binding protein 4 genes were associated with variation in backfat thickness, wereas no association was found for the SNPs of the insulin-like growth factor binding protein 3, peroxisome proliferative active receptor gamma coactivator 1A, proteasome 26S subunit ATPase 1 and corticotropin releasing hormone genes. Additionally, two chromosomal regions of BTA14 were associated with the trait in this work. / A raça Canchim tem sido utilizada na bovinocultura de corte como alternativa de intensificação de produção. Grupos genéticos Canchim (5/8 Charolês + 3/8 Zebu) e MA (descendentes de touros Charoleses e de 1/2 Canchim + 1/2 Zebu) constituem essa raça sintética, que tem bom potencial de crescimento e adaptação tropical, mas deposita pouca gordura quando alimentada à pasto. Espessura de gordura subcutânea, quantidade e distribuição de gordura total têm grande impacto sobre a qualidade da carne e carcaça em gado de corte. Por este motivo, pesquisas têm sido realizadas com o objetivo de aumentar a deposição de gordura nessa raça, incluindo a busca por marcadores moleculares que auxiliem a identificação de animais com maior potencial genético para deposição de gordura. Para que a genética molecular possa ser incorporada nos programas de melhoramento de bovinos de corte, é essencial que a associação entre marcadores moleculares e características de produção seja avaliada na população em que se deseja utilizá-los. Há relatos de muitos genes candidatos e segmentos cromossômicos associados com a variação da deposição de gordura em bovinos. O objetivo desse trabalho foi identificar SNPs associados à deposição de gordura em animais da raça Canchim criados à pasto. Para tanto, foram prospectados SNPs nos genes do fator de aumento de desenvolvimento e de diferenciação 1 e proteína ligante ao fator de crescimento semelhante à insulina 3, verificada a a associação de SNPS presentes nos genes da proteína ligante do fator de crescimento semelhante à insulina 3, coativador de proliferação de peroxissomo ativo por receptor gama 1A, subunidade 26S ATPse do proteassomo, do fator de aumento de desenvolvimento e de diferenciação 1, hormônio liberador de corticotrofina e proteína ligante de ácido graxo 4 com espessura de gordura em animais Canchim e MA. Além disso, a existência de regiões genômicas associadas com espessura de gordura subcutânea foi investigada em populações Canchim e MA usando um chip de 54 K em fenótipos extremos. Dentre as regiões com associação significativa, foram eleitas aquelas presentes no BTA14 para validar a associação por análise de haplótipos na população toda. SNPs nos genes do fator de aumento de desenvolvimento e de diferenciação 1 e da proteína ligante de ácido graxo 4 foram associados à variação de espessura de gordura subcutânea, enquanto que nenhuma associação foi observada para os SNPs dos genes proteína ligante ao fator de crescimento 7 semelhante à insulina 3, coativador de proliferação de peroxissomo ativo por receptor gama 1A, subunidade 26S ATPse do proteassomo e hormônio liberador de corticotrofina. Também foram associadas com a característica avaliada nesse trabalho, duas regiões cromossômicas do BTA14.

Introdução: O número de estudos relacionados à Síndrome Metabólica (SM) vem aumentando nos últimos anos, muitas vezes motivados pelo aumento do número de casos de sobrepeso/obesidade e diabetes Tipo II levando ao desenvolvimento de doenças cardiovasculares e, como consequência, infarto agudo do miocárdio e AVC, dentre outros desfechos desfavoráveis. A SM é uma doença multifatorial composta de cinco características, porém, para que um indivíduo seja diagnosticado com ela, possuir pelo menos três dessas características torna-se condição suficiente. Essas cinco características são: Obesidade visceral, caracterizada pelo aumento da circunferência da cintura, Glicemia de jejum elevada, Triglicérides aumentado, HDL-colesterol reduzido, Pressão Arterial aumentada. Objetivo: Estabelecer a rede de associações entre os fenótipos que compõem a Síndrome Metabólica através do aprendizado de estruturas de dependência, decompor a rede em componentes de correlação genética e ambiental e avaliar o efeito de ajustes por covariáveis e por variantes genéticas exclusivamente relacionadas à cada um dos fenótipos da rede. Material e Métodos: A amostra do estudo corresponderá a 79 famílias da cidade mineira de Baependi, composta por 1666 indivíduos. O aprendizado de estruturas de redes será feito por meio da Teoria de Grafos e Modelos de Equações Estruturais envolvendo o modelo linear misto poligênico para determinar as relações de dependência entre os fenótipos que compõem a Síndrome Metabólica / Introduction: The number of studies related to Metabolic Syndrome (MetS) has been increasing in the last years, encouraged by the increase on the overweight / obesity and Type II Diabetes cases, leading to the development of cardiovascular disease and, therefore, acute myocardial infarction and stroke, and others unfavorable outcomes. MetS is a multifactorial disease containing five characteristics, however, for an individual to be diagnosed with MetS, he/she may have at least three of them. These characteristics are: Truncal Obesity, characterized by increasing on the waist circumference, increasing on Fasting Blood Glucose, increasing on Triglycerides, decreasing on HDL cholesterol and increasing on Blood Pressure. Aims: Establish the best association network between MetS phenotypes through structured dependency learning between phenotypes considering genetic variants exclusively related to each phenotype. Materials and Methods: The study sample is composed of 79 families, 1666 individuals of a city in a rural area of Brazil, called Beapendi. Structured learning will use graph theory and Structural Equations Models to establish the dependency relations between MetS phenotypes

Microarray technology is a promising approach for the simultaneous analysis of multiple single nucleotide polymorphisms (SNPs), which are the most abundant form of genetic variation. In this thesis enzyme-assisted microarray-based methods were developed to improve the accuracy and genotype discrimination power of the current methods for SNP genotyping. The improved technology was applied for analysing recessively inherited disease mutations, for Y-chromosomal SNPs in a population study, for an evolutionary analysis of SNPs in flycatchers and for multiplexed quantitative determination of SNP-allele frequencies in pooled DNA samples. A robust attachment chemistry for immobilising oligonucleotides on glass surface was established, based on an evaluation of eight covalent coupling methods. A four-colour fluorescence detection strategy, which enabled a multiplexed quantitative analysis for as little as 2% of a minority allele frequency in pooled samples was generated. Twenty-five Y-chromosomal SNPs were screened in a collection of 300 samples from five Finno-Ugric-speaking populations using minisequencing on microarrays. In these populations six distinct haplotypes were defined by the six SNPs that were polymorphic. Data from five microsatellite markers was combined with the SNP data, revealing shared Y-chromosomal haplotypes between the Finns and the Saami, indicating, in accordance with earlier data, at least two founding Y-chromosomal lineages in these populations. Database screening and subsequent validation of 125 potential SNPs in the highly repetitive type 1 interferon genes and genes coding for proteins in the interferon-related regulatory pathways revealed 25 informative SNPs in the Finnish and Swedish populations. These SNPs were included in a panel for microarray based genotyping that should find a variety of applications in genetic studies due to the important immunoregulatory functions of the IFN family. The significance of sex-chromosome evolution on speciation was investigated in two naturally hybridising flycatcher species (N=459) by analysing a panel of 20 SNPs using minisequencing on microarrays. A strong selection against gene flow across the species boundary of sex-linked genes was observed, as well as a sex-chromosomal influence on male plumage characteristics that have previously been shown to reinforce isolation in these birds. The results suggest a major role for sex-chromosome-mediated isolation of the two flycatcher species.

Η πιο απλή μορφή γενετικής διαφοροποίησης στον άνθρωπο είναι οι μονονουκλεοτιδικοί πολυμορφισμοί (Single Nucleotide Polymorphisms - SNPs). Ο αριθμός αυτού του είδους πολυμορφισμών που έχουν βρεθεί στο ανθρώπινο γονιδίωμα και επηρεάζουν την παραγόμενη πρωτεΐνη αυξάνεται συνεχώς, αλλά η αντιστοίχηση τους σε πιθανές ασθένειες με πειραματικές μεθόδους είναι ασύμφορη από θέμα χρόνου και κόστους. Για αυτό τον λόγο έχουν αναπτυχθεί διάφορες υπολογιστικές μέθοδοι με σκοπό να ταξινομήσουν τους μονονουκλεοτιδικούς πολυμορφισμούς σε παθογόνους και μη. Οι περισσότερες από αυτές τις μεθόδους χρησιμοποιούν ταξινομητές, οι οποίοι παίρνοντας σαν είσοδο ένα σύνολο δομικών, λειτουργικών, ακολουθιακών και εξελικτικών χαρακτηριστικών, επιχειρούν να προβλέψουν αν ένας μονονουκλεοτιδικός πολυμορφισμός είναι παθογόνος ή μη. Για την εκπαίδευση αυτών των ταξινομητών, χρησιμοποιούνται δύο σύνολα μονονουκλεοτιδικών πολυμορφισμών. Το πρώτο αποτελείται από μονονουκλεοτιδικούς πολυμορφισμούς που έχει βρεθεί πειραματικά ότι οδηγούν σε παθογένεια και το δεύτερο από μονονουκλεοτιδικούς πολυμορφισμούς που έχει αποδειχθεί πειραματικά ότι είναι αδρανείς. Οι μέθοδοι αυτές διαφέρουν στα χαρακτηριστικά των μεταλλάξεων που λαμβάνουν υπόψη στην πρόβλεψη τους, καθώς επίσης και στην εκπαίδευση και τη φύση των τεχνικών ταξινόμησης, που χρησιμοποιούν για τη λήψη των αποφάσεων. Το βασικότερο προβλήματα τους ωστόσο έγκειται στο γεγονός ότι καθορίζουν τα χαρακτηριστικά, που θα χρησιμοποιήσουν σαν είσοδο στους ταξινομητές τους με τρόπο εμπειρικό και μάλιστα διαφορετικές μέθοδοι προτείνουν και χρησιμοποιούν διαφορετικά χαρακτηριστικά, χωρίς να τεκμηριώνουν επαρκώς τις αιτίες αυτής της διαφοροποίησης. Δύο ακόμα προβλήματα που δεν έχουν καταφέρει να αντιμετωπίσουν οι υπάρχουσες μεθοδολογίες είναι το πρόβλημα της ανισορροπίας των δύο κλάσεων ταξινόμησης και των ελλιπών τιμών σε πολλά από τα χαρακτηριστικά εισόδου των ταξινομητών, ώστε να επιτυγχάνουν πιο ακριβή και αξιόπιστα αποτελέσματα. Από τα παραπάνω είναι ξεκάθαρο πως υπάρχει μεγάλο περιθώριο βελτίωσης των υπάρχουσων μεθοδολογιών για το συγκεκριμένο πρόβλημα ταξινόμησης. Στην παρούσα διπλωματική εργασία προτείνουμε μια νέα υβριδική μεθοδολογία υπολογιστικής νοημοσύνης, που ξεπερνά πολλά από τα προβλήματα των υπάρχοντων μεθοδολογιών και βελτιώνει με τον τρόπο αυτό την απόδοσή τους. Δύο είναι τα βασικά βήματα που ακολουθήσαμε για την επίτευξη του στόχου αυτού. Πρώτον, συγκεντρώσαμε από τις διαθέσιμες δημόσιες βάσεις δεδομένων, τους μονονουκλεοτιδικούς πολυμορφισμούς που χρησιμοποιήθηκαν για την εκπαίδευση και τον έλεγχο των μοντέλων μηχανικής μάθησης. Συγκεκριμένα, συλλέχθησαν και φιλτραρίστηκαν τα θετικά και αρνητικά σύνολα εκπαίδευσης και ελέγχου, που αποτελούνται από μονονουκλεοτιδικούς πολυμορφισμούς που είτε οδηγούν σε παθογένεια, είτε είναι ουδέτεροι. Για κάθε πολυμορφισμό των δύο συνόλων υπολογίσαμε χρησιμοποιώντας υπάρχοντα διαθέσιμα εργαλεία όσο το δυνατό περισσότερα δομικά, λειτουργικά, ακολουθιακά και εξελικτικά χαρακτηριστικά. Για εκείνα τα χαρακτηριστικά, για τα οποία δεν υπήρχε κάποιο διαθέσιμο εργαλείο υπολογισμού τους, υλοποιήσαμε τον κατάλληλο κώδικα για τον υπολογισμό τους. Το δεύτερο βήμα της διπλωματικής αφορούσε το σχεδιασμό και την υλοποίηση της κατάλληλης υβριδικής μεθόδου για την επίλυση του προβλήματος που μελετάμε. Χρησιμοποιήσαμε μια νέα μέθοδο ταξινόμησης την EnsembleGASVR. Πρόκειται για μια ensemble μεθοδολογία, που συνδυάζει σε ένα ενιαίο πλαίσιο ταξινόμησης οκτώ διαφορετικούς ταξινομητές. Κάθε ένας από αυτούς τους ταξινομητές βασίζεται στον υβριδικό συνδυασμό των Γενετικών Αλγορίθμων και των μοντέλων Παλινδρόμησης Διανυσμάτων Υποστήριξης (nu-Support Vector Regression). Συγκεκριμένα ένας Προσαρμοζόμενος Γενετικός Αλγόριθμος χρησιμοποιείται για να καθοριστεί το βέλτιστο υποσύνολο χαρακτηριστικών, καθώς και οι βέλτιστες τιμές των παραμέτρων των ταξινομητών. Σαν μέθοδο ταξινόμησης των μεταλλάξεων σε ουδέτερες και παθογενείς, προτείνουμε τον nu-SVR ταξινομητή, καθώς παρουσιάζει υψηλή απόδοση, καλή γενίκευση, δεν παγιδεύεται σε τοπικά βέλτιστα, ενώ ταυτόχρονα επιτυγχάνει την ισορροπία μεταξύ της ακρίβειας και της πολυπλοκότητας του μοντέλου. Μάλιστα για να ξεπεράσουμε τα πρόβληματα των ελλιπών τιμών και της ανισορροπίας των δύο κλάσεων ταξινόμησης, αλλά και για να βελτιώσουμε τη συνολική απόδοση της μεθοδολογίας μας, επεκτείναμε τον υβριδικό αλγόριθμο, ώστε να λειτουργεί σαν μία ensemble-συλλογική τεχνική, συνδυάζοντας οκτώ επί μέρους μοντέλα ταξινόμησης. Τα πειραματικά αποτελέσματα της προτεινόμενης μεθοδολογίας ήταν εξαιρετικά ελπιδοφόρα, καθώς η EnsembleGASVR μεθοδολογία υπερτερεί σημαντικά έναντι άλλων ευρέως γνωστών μεθόδων ταξινόμησης παθογενών μεταλλάξεων. / Single Nucleotide Polymorphisms (SNPs) are the most common form of genetic variations in humans. The number of SNPs that have been found in human genome and affect protein functionality is constantly increasing. Finding matches between SNPs and diseases using experimental techniques, is excessive disadvantageous in terms of time and cost. For this reason, several computational methods have been developed. These methods classify polymorphisms as pathogenic and non-pathogenic. Most of them use classifiers, which take as input a set of structural, functional, sequential and evolutionary features and predict whether a single nucleotide polymorphism is pathogenic or neutral. For training these classifiers use two sets of SNPs. The first one consists of SNPs that have been experimentally proven as pathogenic, whereas the second set consists of SNPs that have been experimentally characterized as benign. These methods differ in the classification methods they deploy and in the features they use as inputs. However, the main problem is the determination of an empirically verified set of features for training. Specifically, different methods suggest different feature sets, without adequately documenting the causes of this differentiation. In addition, the existing methodologies do not tackle efficiently the class imbalance problem between positive and negative training sets and the problem of missing values in the datasets. In this thesis a new hybrid computational intelligence methodology is proposed, that overcomes many of the problems of existing methodologies. The proposed method achieves high classification performance and systematizes the selection of relevant features. In the first phase of this study the polymorphisms were gathered from the available public databases and they were used for training and testing of the machine learning models. Specifically, the positive and negative training and test sets were collected and filtered. They consist of single nucleotide polymorphisms that lead to either pathogenesis or are neutral. For each polymorphism of the two sets, using existing available tools, a wide range of structural, functional, sequential and evolutionary features were calculated. For those features for which there was no available tool, the suitable program (code) was developed in order to compute them. In the second step a new embedded hybrid classification method called EnsembleGASVR is designed and implemented. The method uses an ensemble methodology, based on hybrid combination of Genetic Algorithms and nu-Support Vector Regression (nu-SVR) models. An Adaptive Genetic Algorithm is used to determine the optimal subset of features and the optimal values of the parameters of classifiers. We propose the nu-SVR classifier, since it exhibits high performance, good generalization ability, it is not trapped in local optima and achieves a balance between accuracy and complexity of the model. In order to overcome the problem of missing values and class imbalance, we extended the above algorithm to function as a collective ensemble-technique, combining eight individual classification models. In overall, the method achieves 87.45% accuracy, 71.78% sensitivity and 93.16% specificity. These priliminary results are very promising and shows that EnsembleGASVR methodology significantly outperforms other well-known classification methods for pathogenic mutations.

Contexte. Les phénotypes ABO et Rh(D) des donneurs de sang ainsi que des patients transfusés sont analysés de façon routinière pour assurer une complète compatibilité. Ces analyses sont accomplies par agglutination suite à une réaction anticorps-antigènes. Cependant, pour des questions de coûts et de temps d’analyses faramineux, les dons de sang ne sont pas testés sur une base routinière pour les antigènes mineurs du sang. Cette lacune peut résulter à une allo-immunisation des patients receveurs contre un ou plusieurs antigènes mineurs et ainsi amener des sévères complications pour de futures transfusions. Plan d’étude et Méthodes. Pour ainsi aborder le problème, nous avons produit un panel génétique basé sur la technologie « GenomeLab _SNPstream» de Beckman Coulter, dans l’optique d’analyser simultanément 22 antigènes mineurs du sang. La source d’ADN provient des globules blancs des patients préalablement isolés sur papiers FTA. Résultats. Les résultats démontrent que le taux de discordance des génotypes, mesuré par la corrélation des résultats de génotypage venant des deux directions de l’ADN, ainsi que le taux d’échec de génotypage sont très bas (0,1%). Également, la corrélation entre les résultats de phénotypes prédit par génotypage et les phénotypes réels obtenus par sérologie des globules rouges et plaquettes sanguines, varient entre 97% et 100%. Les erreurs expérimentales ou encore de traitement des bases de données ainsi que de rares polymorphismes influençant la conformation des antigènes, pourraient expliquer les différences de résultats. Cependant, compte tenu du fait que les résultats de phénotypages obtenus par génotypes seront toujours co-vérifiés avant toute transfusion sanguine par les technologies standards approuvés par les instances gouvernementales, les taux de corrélation obtenus sont de loin supérieurs aux critères de succès attendus pour le projet. Conclusion. Le profilage génétique des antigènes mineurs du sang permettra de créer une banque informatique centralisée des phénotypes des donneurs, permettant ainsi aux banques de sang de rapidement retrouver les profiles compatibles entre les donneurs et les receveurs. / Background. ABO and Rh(D) phenotyping of both blood donors and transfused patients is routinely performed by blood banks to ensure compatibility. These analyses are done by antibody-based agglutination assays. However, blood is not routinely tested for minor blood group antigens on a regular basis because of cost and time constraints. This can result in alloimmunization of the patient against one or more minor antigens and may complicate future transfusions. Study design and Methods. To address this problem, we have generated an assay on the GenomeLab SNPstream genotyping system (Beckman Coulter, Fullerton, CA) to simultaneously test polymorphisms linked to 22 different blood antigens using donor’s DNA isolated from minute amounts of white blood cells. Results. The results showed that both the error rate of the assay, as measured by the strand concordance rate, and the no-call rate were very low (0.1%). The concordance rate with the actual red blood cell and platelet serology data varied from 97 to 100%. Experimental or database errors as well as rare polymorphisms contributing to antigen conformation could explain the observed differences. However, these rates are well above requirements since phenotyping and cross-matching will always be performed prior to transfusion. Conclusion. Molecular profiling of blood donors for minor red blood cell and platelet antigens will give blood banks instant access to many different compatible donors through the set-up of a centralized data storage system.