Mycoplasma hyopneumoniae é uma bactéria de tamanho diminuto, caracterizada por um genoma pequeno, com baixo conteúdo GC. Está associada com doenças respiratórias de suínos, resultando em prejuízos produtivos e econômicos na indústria animal. A presença de sequências de DNA repetitivas, que ocorrem em grandes quantidades em células eucarióticas, vem sendo cada vez mais identificadas em genomas de procariotos, sendo também associadas a um potencial papel regulador. Uma vez que a regulação da transcrição nesses organismos ainda é pouco entendida, o objetivo do presente estudo foi realizar uma busca in silico por elementos repetitivos nas regiões intergênicas do genoma de M. hyopneumoniae linhagem 7448. Dois tipos de repetições foram selecionados para a busca inicial: tandem e palindromes. Regiões intergênicas de até 500 pb a montante do sítio de início da tradução de todas as CDSs do genoma de M. hyopneumoniae linhagem 7448 foram utilizadas para a predição. Para cada tipo de elemento dois programas computacionais independentes foram utilizados. As predições in silico resultaram em 144 repetições em tandem e 1.171 palindromes. O DNA repetitivo se encontra distribuído a montante de 86% das unidades transcricionais de M. hyopneumoniae linhagem 7448. Análises comparativas entre genomas de micoplasmas demonstraram diferentes níveis de conservação dos elementos repetitivos entre linhagens patogênicas e não-patogênicas. Linhagens patogênicas revelaram uma conservação de 59%, enquanto que a não patogênica, somente de 46%. Através de ensaios de amplificação quantitativa de DNA, foi observado diferentes níveis de expressão em genes codificantes para importantes proteínas, como glicina hidroximetiltransferase, lipoproteína, adesinas e proteína ligadora de GTP. Os genes codificantes para essas proteínas divergiam no número de repetições palindromes e tandens na sua respectiva região intergênica. Além disso, repetições encontradas em 206 genes já descritos como regulados em diferentes condições em M. hyopneumoniae linhagem 232 mostraram aproximadamente 80% de conservação em relação à linhagem M. hyopneumoniae linhagem 7448. Todos esses resultados sugerem um potencial papel regulador das repetições de DNA em tandem e palindromes em Mycoplasma. / Mycoplasma hyopneumoniae is a diminutive bacterium, characterized by a small genome with a low GC content. It is commonly associated with swine respiratory diseases, resulting in productivity and economic losses in the animal industry. Repetitive DNA, which occurs in large quantities in eukaryotic cells, has been increasingly identified in prokaryotic genomes, and has been associated with a potential regulatory function. Once transcription regulation in these organisms is still poorly understood, the aim of the current study was to perform an in silico search of repeat elements in the genomic intergenic regions of M. hyopneumoniae strain 7448. Two types of repeats were selected for initial search: Tandem and Palindromic. Intergenic regions up to 500 bp upstream from start codon of M. hyopneumoniae strain 7448 CDSs were used as input for the software’s prediction. For each type of repeat sequence, two independent software packages were used. Computational analysis results in 144 tandem repeats and 1,171 palindrome elements. The repeats were distributed in the upstream region of 86% of transcriptional units of M. hyopneumoniae strain 7448. Comparative analysis between distinct mycoplasmas, demonstrate different indices of repeat conservation among pathogenic and non-pathogenic strains. Pathogenic strains revealed 59% conservation, while non-pathogenic only 46%. Through assays of quantitative amplification of DNA, different levels of expression in genes coding important proteins have been demonstrated, as glycine hydroxymethyltransferase, lipoprotein, adhesins and GTP-binding protein. These protein coding genes differ in number of palindromes or tandem repeats in respective upstream regions. In addition, repeats found in 206 genes already described to be regulated in different grow conditions in M. hyopneumoniae strain 232 showed almost 80% of conservation in relation to M. hyopneumoniae strain 7448. All these findings, suggests a potential regulatory role of tandem and palindrome DNA repeats.

De nos jours, l’augmentation du volume des données de séquençage est bien plus forte que celle de notre capacité à analyser ces données. En lien avec ce déluge de données et le besoin urgent de nouveaux outils bioinformatiques pour les analyser, notre travail consiste à développer de nouveaux algorithmes pour mieux comprendre les relations entre séquence, structure, et fonction des protéines. Les protéines contiennent de larges portions de séquences périodiques, qui forment des motifs d’acides aminés répétés les uns à la suite des autres que l’on appelle des répétitions en tandem. Elles se retrouvent dans 14% des protéines. De nombreuses études ont montré leur importance fonctionnelle ainsi que leur implication dans de nombreuses maladies humaines, notamment le cancer. Ici, nous montrons l’importance d’adopter une approche incluant plusieurs outils de détection de répétition en tandem afin de s’assurer d’obtenir le jeu de données le plus complet. Nous avons ainsi réalisé un pipeline approprié, et développé deux outils spécifiques : un filtre, pour gagner en rapidité, et un score, pour sélectionner les répétitions les plus pertinentes dans les régions structurées des protéines. Enfin, nous avons utilisé ce pipeline sur une sélection de 94 protéomes. Cette analyse a permis de mettre à jour le précédent recensement des répétitions, montrant que 64% des protéines contenaient des répétitions en tandem. Elle a également permis de mieux comprendre les répétions en tandem dans leurs caractéristiques, leurs compositions et leurs implications dans les maladies humaines. / Today, the growth of protein sequencing data significantly exceeds the growth of capacities to analyze these data. In line with this data deluge and urgent needs in new bioinformatics tools our work deals with the development of new algorithms to better understand the sequence-structure-function relationship. Proteins contain a large portion of periodic sequences representing arrays of repeats that are directly adjacent to each other, so called tandem repeats (TRs). TRs occur at least in 14% of all proteins. Highly divergent, they range from a single amino acid repetition to domains of 100 or more repeated residues. Numerous studies demonstrated the fundamental functional importance of such TRs and their involvement in human diseases, especially cancers. Here we show the importance of integrating several TR detectors to get the most complete set of TRs in proteomes. We designed an appropriate pipeline and developed a filter to speed the process as well as a new scoring module to select relevant structured TRs. In addition, we undertook a large scale analysis of TRs in 94 proteomes. This large scale analysis allowed us to update previous census of TR showing that TRs occurs in 64% of all proteins and leads to a better understanding of TR in terms of their characteristics, composition and implication in human disease.

Apart from contributing to Computer Science, this research also contributes to Bioinformatics, a subset of the subject discipline Computational Biology. The main focus of this dissertation is the development of a data-analytical and theoretical algorithm to contribute to the analysis of DNA, and in particular, to detect microsatellites. Microsatellites, considered in the context of this dissertation, refer to consecutive patterns contained by genomic sequences. A perfect tandem repeat is defined as a string of nucleotides which is repeated at least twice in a sequence. An approximate tandem repeat is a string of nucleotides repeated consecutively at least twice, with small differences between the instances. The research presented in this dissertation was inspired by molecular biologists who were discovered to be visually scanning genetic sequences in search of short approximate tandem repeats or so called microsatellites. The aim of this dissertation is to present three algorithms that search for short approximate tandem repeats. The algorithms comprise the implementation of finite automata. Thus the hypothesis posed is as follows: Finite automata can detect microsatellites effectively in DNA. "Effectively" includes the ability to fine-tune the detection process so that redundant data is avoided, and relevant data is not missed during search. In order to verify whether the hypothesis holds, three theoretical related algorithms have been proposed based on theorems from finite automaton theory. They are generically referred to as the FireìSat algorithms. These algorithms have been implemented, and the performance of FireìSat2 has been investigated and compared to other software packages. From the results obtained, it is clear that the performance of these algorithms differ in terms of attributes such as speed, memory consumption and extensibility. In respect of speed performance, FireìSat outperformed rival software packages. It will be seen that the FireìSat algorithms have several parameters that can be used to tune their search. It should be emphasized that these parameters have been devised in consultation with the intended user community, in order to enhance the usability of the software. It was found that the parameters of FireìSat can be set to detect more tandem repeats than rival software packages, but also tuned to limit the number of detected tandem repeats. Copyright / Dissertation (MSc)--University of Pretoria, 2010. / Computer Science / unrestricted

Plants have evolved many mechanisms to defend against herbivores and pathogens. In many cases, these mechanisms took other duties. One example of such a neofunction- alisation would be carnivory. Carnivory evolved from the defence against herbivores. Instead of repelling the predator with a bitter taste, the plant kills it and absorbs its nutrients. A second example can be found in the pollination process. Many of the genes involved here were originally part of defence mechanisms against pathogens. In this thesis, I study these two examples on a genomic and transcriptomic level. The first project, Genomics of carnivorous Droseraceae, aims at obtaining annotated genome sequences of three carnivorous plants. I assembled the genome of Aldrovanda vesiculosa, annotated those of A. vesiculosa, Drosera spatulata and Dionaea muscipula and com- pared their genomic contents. Because of the high repetitiveness of the D. muscipula genome, I also developed reper, an assembly free method for detection, classification and quantification of repeats. With that method, we were able to study the repeats without the need of incorporating them into a genome assembly. The second large project investigates the role of DEFL (defensin-like) genes in pollen tube guidance in tobacco flowers. We sequenced the transcriptome of the SR1 strain in different stages of the pollination process. I assembled and annotated the transcriptome and searched for differentially expressed genes. We also used a method based on Hidden- Markov-Models (HMM) to find DEFLs, which I then analysed regarding their expression during the different stages of fertilisation. In total, this thesis results in annotated genome assemblies of three carnivorous Droser- aceae, which are used as a foundation for various analyses investigating the roots of car- nivory, insights into the role of DEFLs on a transcriptomic level in tobacco pollination and a new method for repeat identification in complex genomes. / Im Laufe der Evolution haben Pflanzen viele Methoden entwickelt, um sich gegen Fress- feinde und Pathogene zu verteidgen. Viele dieser Methoden wurden im Laufe der Zeit umfunktioniert. Ein Beispiel hierfür ist die Karnivorie, welche aus der Verteidigung ge- gen Fressfeinde entstanden ist. Anstelle einen Angreifer durch bitteren Geschmack zu vertreiben, tötet die Pflanze das Tier und nimmt seine Nährstoffe auf. Ein weiteres Bei- spiel ist der Bestäubungs- und Befruchtungsprozess. Viele der Gene, die hier involviert sind, stammen ursprünglich aus Mechanismen zur Verteidigung gegen Pathogene. In dieser Arbeit untersuche ich diese beiden Beispiele auf genomischer und transkrip- tomischer Ebene. Die Zielsetzung des ersten Projekts, Genomik von karnivoren Dro- seraceaen, ist es, assemblierte und annotierte Genome von drei karnivoren Pflanzen zu generieren. Ich habe dazu das Genom von Aldrovanda vesiculosa assembliert und dieses, sowie die Genome von Drosera spatulata und Dionaea muscipula annotiert und mit- einander verglichen. Aufgrund des hohen Anteils repetitiver Elemente im D. muscipula Genom habe ich reper, eine Methode zum Detektieren, Klassifizieren und Quantifizieren von Repeats, entwickelt. Mit dieser Methode ist es nun möglich, repetitive Elemente zu untersuchen, ohne diese in einem Genomassembly integrieren zu müssen. Das zweite große Projekt untersucht die Rolle von DEFL (defensin-like) Genen im Pollenschlauchwachstum in Tabakblüten. Dazu haben wir das Transkriptom der SR1 Variante zu verschiedenen Zeitpunkten im Befruchtungsprozess sequenziert. Ich habe dieses Transkriptom assembliert und annotiert und darin nach differentiell exprimierten Genen gesucht. Zudem haben wir mit einer auf Hidden Markov Modellen (HMM) ba- sierten Methode nach DEFL Genen gesucht und ich habe die Expression dieser in den verschiedenen Stadien untersucht. Zusammenfassend beinhalten die Ergebnisse dieser Thesis annotierte Genomassemb- lies von drei karnivoren Droseraceaen, Erkenntnisse über die Rolle von DEFL Genen bei der Befruchtung auf einer transkriptomischen Ebene und eine neue Software zur Analyse von repetitiven Elementen in komplexen Genomen.

Selectins are key adhesion molecules responsible for initiating a multistep process that leads a cell out of the blood circulation and into a tissue or organ. Their extracellular structure is composed of an N-terminal extracellular C-type lectin like domain, followed by an Endothelial Growth Factor like domain (EGF), a defined number of short consensus repeats SCR. The adhesion of cells (expressing ligands) to the endothelium (expressing the selectin i.e., E-selectin) occurs through spatio-temporally regulated interactions that are mediated by multiple intra- and inter-cellular components. Furthermore, selectins play a role beyond fixing cells to a specific location by regulating important signaling pathways in the migrating cell during physiological and pathological processes. These interactions start mainly with the binding of the lectin domain of selectins and ligand on cells. Therefore, structural/functional studies to date have mainly focused on the direct interactions of the lectin domain of E-selectin with its ligands while other domains and conformational dynamics received less attention. For this purpose, we produced a number of different recombinant E-selectin proteins with and without artificial oligomerization and with changes in the SCR units in addition to proteins where strategic residues will be mutated to change the conformation of the selectin to an extended conformer. Moreover, double cysteine mutant candidates were produced for maleimide labeling for the real-time SM-FRET (single molecule fluorescence resonance energy transfer) studies to assess conformational dynamics of E-selectin. Using a comprehensive set of static- and flow-based assays, we concluded that SCR domains play a role by enhancing the interaction of recombinant E-selectin proteins with E-selectin ligand, while dimerization and extension of the lectin domain improve the binding. However, our double cysteine mutants purification and labeling requires further optimization to be utilized to study the conformational dynamics of E-selectin binding to its ligands using SM-FRET and force microscopy. Furthermore, our experiments extend to highlight the importance of phosphatases in regulating signaling pathways that are affected by E-selectin binding to migrating cells. Collectively, these studies are beneficial to understand the mechanistic details of cell adhesion and migration of cells using the established model system of hematopoietic stem cells (HSCs) adhesion to the selectin expressing endothelial cells.

Over half the human genome consists of repetitive sequences. One major class is the tandem repeats (TRs), which are defined by their location in the genome, repeat unit, and copy number. TRs loci that exhibit variant copy numbers are called Variable Number Tandem Repeats (VNTRs). High VNTR mutation rates of approximately 0.0001 per generation make them suitable for forensic studies, and of interest for potential roles in gene regulation and disease. TRs are generally divided into three classes: 1) microsatellites or short tandem repeats (STRs) with patterns <7 bp; 2) minisatellites with patterns of seven to hundreds of base pairs; and 3) macrosatellites with patterns of >100 bp. To date, mini- and macrosatellites have been poorly characterized, mainly due to a lack of computational tools. In this thesis, I utilize a tool, VNTRseek, to identify human minisatellite VNTRs using short-read sequencing data from nearly 2,800 individuals and developed a new computational tool, MaSUD, to identify human macrosatellite VNTRs using data from 2,504 individuals. MaSUD is the first high-throughput tool to genotype macrosatellites using short reads. I identified over 35,000 minisatellite VNTRs and over 4,000 macrosatellite VNTRs, most previously unknown. A small subset in each VNTR class was validated experimentally and in silico. The detected VNTRs were further studied for their effects on gene expression, ability to distinguish human populations, and functional enrichment. Unlike STRs, mini- and macrosatellite VNTRs are enriched in regions with functional importance, e.g., introns, promoters, and transcription factor binding sites. A study of VNTRs across 26 populations shows that minisatellite VNTR genotypes can be used to predict super-populations with >90% accuracy. In addition, genotypes for 195 minisatellite VNTRs and 22 macrosatellite VNTRs were shown to be associated with differential expression in nearby genes (eQTLs). Finally, I developed a computational tool, mlZ, to infer undetected VNTR alleles and to detect false positive predictions. mlZ is applicable to other tools that use read support for predicting short variants. Overall, these studies provide the most comprehensive analysis of mini- and macrosatellites in human populations and will facilitate the application of VNTRs for clinical purposes.

Monoploid potato genotypes represent plant material that is free from the "genetic load" of lethal and severely deleterious alleles normally present in the highly heterozygous cultivated potato species. Field evaluations enabled the identification of agronomically superior monoploid potato genotypes from a population of more than 100 anther-derived monoploids. Chemical fusion and electrofusion between pairs selected from 31 superior monoploids resulted in the production of three different groups of intermonoploid somatic hybrids. The hybridity of somatic hybrid plants and calluses was confirmed through PCR-based amplification of simple sequence repeat (SSR) sequences in the potato genome. Polymorphic SSR loci between the monoploid parents of a particular group of somatic hybrids were used to separate true somatic hybrids (heterozygous at the loci) from parental somaclones regenerating from unfused protoplasts (homozygous for one parental band at the loci). One group of somatic hybrids (SH1, SH2 and SH2B) was of particular interest because it resulted from the fusion of a S. phureja monoploid to a high acetylleptinidine-producing monoploid derived from an F1 hybrid between S. chacoense and S. phureja. The leptine acetylleptinidine (ALD) is produced only by some accessions of S. chacoense Bitt. and provides resistance to feeding by the Colorado potato beetle (Leptinotarsa decemlineata Say) when present in sufficient concentrations. The somatic hybrids produced moderate levels of ALD in leaves and stems (roughly 60% that of a high ALD-producing S. chacoense clone). Pollinations of SH1, SH2 and SH2B by several diploid and tetraploid potato clones resulted in three fruit on SH2, one fruit on SH2B and no fruit on SH1. Two resulting progeny populations of SH2 [SH2A = SH2 × S. andigena 8-1 (4x); SH2P = SH2 × S. phureja 66AP11-53 (2x)] expressed higher fertility than the original somatic hybrids and were sexually crossed as both male and female parents to S. tuberosum cv. Atlantic. All of the SH2 progeny populations expressed acetylleptinidines, albeit at lower levels than the SH2 somatic hybrid, providing strong evidence that the genes controlling acetylleptinidine production are dominant. Variation for ALD expression in the SH2 progeny indicated one or a few genes with additive effect controlling the ALD trait. In addition, the choice of male parent in sexual crosses to SH2 affected subsequent ALD expression in progeny populations. The SH2 progeny represent an important first step towards transferring acetylleptinidines to cultivated potato. SH1, SH2 and SH2B appeared to be negatively affected by an unusually high ploidy (hexaploid, 6x). Field-grown plants produced many tubers (mean = 35) of low weight (mean = 10.4 g) and were stunted in appearance. Anther culture of SH2 yielded triploid regenerants (3x). These regenerants may be more phenotypically normal than the original somatic hybrids because of lower ploidy. Segregation of SSR alleles in the triploid anther culture regenerants provided evidence that the hexaploid somatic hybrid SH2 genome is comprised of four homologous genomes of CP2-103 (the high leptine-producing monoploid) and two homologous genomes of 13-14 203 (the S. phureja monoploid). / Master of Science

Submitted to the faculty of the School of Informatics in partial fulfillment of the requirements for the degree Master of Science in Bioinformatics in the School of Informatics,Indiana University July, 2004 / DNA markers have revolutionized the field of genetics by increasing the pace of genetic analysis. Simple sequence repeats (SSRs) are repetitions of nucleotide motifs of 1 to 5 bases and are currently the markers of choice in many plant and animal genomes due to their abundant distribution in the genomes, hypervariable nature and suitability for high-throughput analysis. While SSRs, once developed, are extremely valuable, their development is time consuming, laborious and expensive. Sequences from many genomes are continuously made freely available in the public databases and mining of these sources using computational approaches permits rapid and economical marker development. Expressed sequence tags (ESTs) are ideal candidates for mining SSRs not only because of their availability in large numbers but also due to the fact that they represent expressed genes. Large scale SSR mining efforts in plants to date focused on monocotyledonous plants. In this project, an efficient SSR identification tool was developed and used to mine SSRs from more than 53 dicotyledonous species. A total of 92,648 non-redundant ESTs or 6.0% of the 1.54 million dicotyledonous ESTs investigated in this study were found to contain SSRs. The frequency of non-redundant-ESTs containing SSRs among the species investigated ranged from 2.65% to 16.82%. More than 80% of the non-redundant ESTs having SSRs contained a single SSR repeat while others contained 2 or more SSRs. An extensive analysis of the occurrence and frequencies of various SSR types revealed that the A/T mononucleotide, AG/GA/CT/TC dinucleotide, AAG/AGA/GAA/CTT/TTC/TCT trinucleotide and TTTA and TTAA tetranucleotide repeats are the most abundant in dicotyledonous species. In addition, an analysis of the number of repeats across species revealed that majority of the mononucleotide SSRs contained 15-25 repeats while majority of the di- and tri-nucleotide SSRs contained 5-10 repeats. By providing valuable information on the abundance of SSRs in ESTs of a large number of dicotyledonous species, this study demonstrates the potential of computational mining approach for rapid discovery of SSRs towards the development of markers for genetic analysis and related applications.

Amyloid aggregation involves the spontaneous formation of fibers from misfolded proteins. This process requires low energy input, results in robust fibers, and is thus of interest from a materials manufacturing perspective. The effect of glutamine content and hydrophobicity of template peptides on amyloid aggregation of a template-peptide system involving myoglobin was studied at near-physiological conditions by Fourier transform infrared spectroscopy, atomic force microscopy, field emission scanning electron microscopy, and nanoindentation. Hydrophobic interactions were found to be important for controlled hierarchical fiber growth via a cooperative mechanism, with the largest effect in myoglobin mixtures. Hydrophobic packing increased for most systems as aggregation progressed. The largest changes in structure occurred upon drying. When myoglobin was present with the highest glutamine-containing template (P7), the high glutamine peptide was not effective as a template, since it appeared to prefer self-catalysis. A low level of glutamine in some unordered templates was insufficient for amyloid development. However, templating was more important in glutamine-free templates mixed with myoglobin, which formed fibers with a surprisingly high elastic modulus. This may have been due to template patterning. Nanoindentation results confirmed that glutamine blocks were not necessary for strong intermolecular interactions and cooperative fibril formation. / Master of Science

No description available.