The real estate sector in Sweden accounts for a significant share of energy consumption andgreenhouse gas emissions in society. The increased electrification, driven by factors such asdigitalization and the use of electric cars, further contributes to the industry's climate impact.However, there are opportunities for property owners to effectively manage electricityconsumption and reduce the negative climate impact. Managing power peaks, which occurduring periods of high electricity consumption, is crucial to reduce strain on the power gridand the use of fossil fuels. It is also a key factor in achieving international sustainability goalssuch as Agenda 2030 and the Paris Agreement. Reducing peak loads can also lead to lowerelectricity costs for buildings. The purpose of this study has been to investigate the challenges and opportunities for reducingpeak power demand at KTH Campus in Östermalm, Stockholm, and to examine theenvironmental and economic benefits that can be achieved through this. Using data providedby the real estate company Akademiska Hus, an overview of the overall electricity consumptionat KTH Campus was conducted. In addition, potential measures to reduce peak power demandand finding common incentives for tenants and property owners for implementing suchinvestments were investigated through a qualitative study. The results of the study show that there are measures that property owners can take to reducepower peaks. The suggested measures include both technical investments and influencing andchanging tenants' electricity consumption. The analysis of electricity usage for the study objectrevealed that the hour that primarily should be assessed is 12:00 PM, when the highest numberof power peaks occur. Additionally, potential measures such as upgrading ventilation systemsand optimising the use of laboratory fume hoods were identified to reduce electricityconsumption and, in turn, power peaks at KTH Campus. Improved data utilisation andtransparent knowledge sharing between tenants and property owners can be key tosuccessfully reducing power peaks. Challenges in implementing the proposed measuresinvolve changing tenants' behaviour and managing split incentives between the landlord andtenant. The focus of this study was to analyse existing data on power load distribution andcomprehend it through interviews with experts within the field. Another way of conducting asimilar type of study on how to reduce power peaks could be to develop different strategiesfor analysing data or gathering alternative data. / Den svenska fastighetssektorn står för en betydande andel av energiförbrukningen ochutsläppen av växthusgaser i samhället. Ökad elektrifiering, drivet av faktorer som digitaliseringoch användning av elbilar, bidrar ytterligare till branschens klimatpåverkan. Det finnsemellertid möjligheter för fastighetsägare att effektivt hantera elförbrukningen och minskaden negativa klimatpåverkan. Att hantera effekttoppar, som uppstår under perioder med högelförbrukning, är avgörande för att minska belastningen på elnätet och användningen avfossila bränslen. Det är också en viktig faktor för att uppnå internationella hållbarhetsmål somAgenda 2030 och Parisavtalet. Minskade effekttoppar kan även leda till lägre elkostnader förbyggnader. Syftet med denna studie har varit att undersöka utmaningar och möjligheter att minskaeffekttoppar i elförbrukningen på KTH Campus på Östermalm, Stockholm, och att undersökade miljömässiga och ekonomiska fördelarna som kan uppnås genom detta. Med hjälp av datafrån fastighetsbolaget Akademiska Hus genomfördes en kartläggning av den totalaelförbrukningen vid KTH Campus. Dessutom undersöktes potentiella åtgärder för att minskaeffekttoppar och att hitta gemensamma incitament för hyresgäster och fastighetsägare attimplementera sådana investeringar genom en kvalitativ studie. Resultaten av studien visar att det finns åtgärder som fastighetsägare kan vidta för att minskaeffekttopparna. De föreslagna åtgärderna inkluderar både tekniska investeringar och påverkansamt förändring av hyresgästers elförbrukning. Analysen av elförbrukningen för studieobjektetvisade att timmen som i huvudsak bör beaktas är kl. 12:00, då flest effekttoppar förekommer.Dessutom identifierades potentiella åtgärder som uppgradering av ventilationssystem ochoptimering av användningen av dragskåp i labb för att minska elförbrukningen och därmedäven effekttopparna vid KTH Campus. Förbättrad användning av data och transparentkunskapsdelning mellan hyresgäster och fastighetsägare är också potentiella lösningar för attfrämja investeringar i energieffektivitet. Utmaningar med att implementera de föreslagnaåtgärderna innefattar att ändra hyresgästers beteende och hantera delade incitament mellanfastighetsägare och hyresgäster. Fokus för denna studie var att analysera data över effektförbrukningen på KTH och skapaförståelse genom intervjuer med kunniga inom området. Ett annat sätt att genomföra enliknande studie om hur man minskar effekttoppar skulle kunna vara att utveckla olikastrategier för att analysera data eller samla alternativ data.

Denna studies syfte är att testa om det finns ett statistiskt samband mellan det nominella aktiepriset och ex-dagseffekten på Nasdaq OMX Stockholm. Ett tydligt samband skulle därmed vara ett ytterligare motiv till företagens beslut om genomförandet av aktiesplit för att revidera aktiens nominella pris. Studiens hypotes lyder därför att det finns ett negativt samband mellan det nominella aktiepriset och ex-dagseffekten, som visats i tidigare studie på den amerikanska börsen NYSE. Studien har genomförts i positivistisk tradition genom statistiska analyser och tester för att klargöra ett eventuellt samband mellan den beroende variabeln ex-dagseffekten och den oberoende variabeln nominella priset. All empirisk data har hämtats från databasen Thomson Reuter Datastream, sammanställts i Excel kalkylblad, analyserats i statistikprogrammet MiniTab och redovisats i två uppsättningar. Studiens resultat visar inget samband mellan det nominella priset och ex-dagseffekten under perioden 2011 till 2015. Nollhypotesen kan inte förkastas och resultaten indikerar försumbar korrelation och förklaringsgrad genom regression. Resultatet är annorlunda från en tidigare studie som konstaterat ett tydligt samband mellan samma variabler på börsen i USA. Det teoretiska bidraget består främst av besvarandet av studiens syfte där det nominella prisets betydelse ter sig annorlunda på den svenska marknaden mot den amerikanska. Det praktiska bidraget från studien ger företagsledare för börsnoterade bolag samt fondbolag och aktörer på den finansiella marknaden en utökad kunskap om rådande förhållanden på marknaden för att förbättra beslutsunderlaget vid eventuella aktiesplittar eller investeringar. Som förslag till fortsatt forskning uppmuntras det att undersöka huruvida det nominella prisets betydelse skiljer sig mellan olika marknader. Förslagsvis kan framtida studier mäta effektiviteten på stockholmsbörsen på dagen för aktiesplit som också i teorin är en mätbar händelse på de finansiella marknaderna under rätt förutsättningar. / The aim of this study is to test for a correlated connection between the nominal stockprice and the price-drop-to-dividend ratio on the Swedish stock market Nasdaq OMX Stockholm. A strong correlated connection would be another motive for company managers to implement a stock split to reduce the nominal stock price. Therefore the hypothesis of the study is that there is a negative correlation between the two variables, just as shown in a recent study on the American stockmarket NYSE. This study has been computed with a positivistic approach through statistical tests and analysis to discover an eventual correlated connection between the dependent variable price-drop-to-dividend ratio and the independent variable nominal price. All empirical data was collected from Thomson Reuter Datastream, compiled in Excel worksheet, analyzed with statistical software MiniTab and presented in two sets of data. The result of this study shows no correlated connection between the nominal stock price and the pricedrop-to-dividend ratio during the period of 2011 to 2015. The null hypothesis can not be rejected and the results of the analysis indicate negligible correlation and coefficient of determination through regression, regardless which sets of data observed. The result is different to a recent study which has shown a significant correlated connection between the same two variables on the American stock market NYSE. The theoretical contribution comprises foremost of answering the aim of the study where the nominal prices impact acts differently on the Swedish stock market compared to the American. Also a presenting of the mean value of price-fall-to-dividend ratio for the period examined is a theoretical contribution. The practical contribution from this study give managers for listed companies along with fund managers and operators on the financial markets an increased knowledge about current influences on the market which improves their ability to make decisions about stock split and future investments. For future studies we suggest to do more research on how the impact of nominal prices differ among markets. Tentatively future research can measure the stockholm market efficiency on the day of stocksplit which according to theory is another measureable event on the financial markets under the right circumstances.

Our sample (Xit; Yit) consists of pairs of variables. The real variable Xit measures the creditworthiness of individual i in period t. The Bernoulli variable Yit is the default indicator of individual i in period t. The objective is to estimate a credit rating system, i.e. to particularly divide the range of the creditworthiness into several rating classes, each with a homogeneous default risk. The field of change point analysis provides a way to estimate the breakpoints between the rating classes. As yet, the literature only considers models without dependencies or with dependence only in cross section. This contribution proposes multi-period models including dependencies in cross section as well as in longitudinal section. Furthermore, estimators for the model parameters are suggested. The estimators are applied to a data set of a German credit bureau.

Les aminoacyl-ARNt synthétases catalysent la formation des aminoacyl-ARNt, utilisés lors de la synthèse protéique et peuvent également former des complexes multi-synthétasiques (MSC). Chez S. cerevisiae, le complexe AME associe les glutamyl- et méthionyl-ARNt synthétases à la protéine d’ancrage Arc1 et joue un rôle primordial dans la coordination de l’expression des génomes nucléaire et mitochondrial. Tous les composants de ce MSC sont multi-localisés et assurent des fonctions essentielles dans d’autres compartiments. Pour étudier ces localisations multiples, nous avons élaboré un outil, basé sur la Split-GFP, qui nous permet de visualiser spécifiquement la fraction organellaire d’une protéine multi-localisée. Pour cela, la GFP a été séparée en deux fragments : i) β1-10, restreint à un compartiment subcellulaire et ii) β11, fusionné aux protéines d’intérêts. Cet outil nous a permis d’étudier diverses relocalisations, ainsi que de délimiter des signaux d’import. / Aminoacyl-tRNA synthetases catalyze aminoacyl-tRNA formation, required for protein synthesis but can also associate into multi-synthetase complexes (MSC). In S. cerevisiae, the AME complex contains glutamyl- and methionyl-tRNA synthetases bound to the anchor protein Arc1 and is responsible for the coordination of nuclear and mitochondrial genome expression. The three MSC partners are multi-localized and present simultaneously in several compartments. The detection of the organellar pools of these multilocalized proteins in vivo is difficult, since they are mainly cytosolic. Therefore, we engineered a split-GFP based localization tool that allows us to specifically visualize organellar fractions of multi-localized proteins. To do so, GFP was split into two parts: β1-10, restricted to a subcellular compartment and β11, fused to the protein of interest. This tool allowed us to study relocalization of cytosolic proteins and characterize targeting signals.

Alle Organismen sind für ihr Überleben auf Metalle angewiesen. Hierbei gibt es für jedes Metall einen Konzentrationsbereich, der das Optimum zwischen Metallmangel, -bedarf und -toxizität darstellt. Es gilt mittlerweile als erwiesen, dass alle Organismen zur Aufrechterhaltung des Metallgleichgewichts ein komplexes Netzwerk von Proteinen und niedermolekularen Verbindungen entwickelt haben. Die molekularen Komponenten dieses Netzwerks sind nur zu einem Teil bekannt und charakterisiert: In den letzten Jahren wurden einige Proteinfamilien identifiziert, deren Mitglieder Metalle durch Lipidmembranen transportieren. Eine dieser Metalltransporterfamilien ist die Cation Diffusion Facilitator (CDF)-Familie: Alle charakterisierten Mitglieder exportieren Metalle aus dem Zytoplasma – entweder in zelluläre Kompartimente oder aus der Zelle heraus. Von den zwölf Mitgliedern dieser Familie in Arabidopsis thaliana (A. thaliana) – Metall Toleranz Protein (MTP)-1 bis -12 – wurden bisher AtMTP1 und AtMTP3 charakterisiert. In dieser Arbeit wird die Charakterisierung von AtMTP2 beschrieben. Wie die homologen Proteine AtMTP1 und AtMTP3 führt AtMTP2 zu Zn-Toleranz, wenn es heterolog in Zn-sensitiven Hefemutanten exprimiert wird. Mit AtMTP2 transformierte Hefemutanten zeigten darüber hinaus erhöhte Co-Toleranz. Expression von chimären AtMTP2/GFP Fusionsproteinen in Hefe, A.thaliana protoplasten und in stabil transformierten A.thalinana Planzenlinien deutet auf Lokalisation of AtMTP2 in Membranen des Endoplasmatischen Retikulums (ER) hin, wenn GFP an den C-Terminus von MTP2 fusioniert wird. Fusion of GFP an den N-Terminus von AtMTP2 führte zu Lokalisation in der vakuolären Membran, was wahrscheinlichsten auf Fehllokalisierung durch Maskierung eines ER-Retentionsmotivs (XXRR) am N-Terminus von AtMTP2 zurückgeht. Dies legt nahe, dass AtMTP2 die erwähnten Metalle in das Endomembransystem der Zelle transportieren kann. Eine gewebespezifische Lokalisierung wurde mit Pflanzen durchgeführt, die das β-Glucuronidase (GUS)-Reporterprotein bzw. chimäre Fusionsproteine aus EGFP und AtMTP2 unter Kontrolle des nativen pMTP2-Promotors exprimierten. Diese Experimente bestätigten zum einen, dass der pMTP2-Promotor nur unter Zn-Defizienz aktiv ist. GUS-Aktivität wurde unter diesen Bedingungen in zwei Zonen der Wurzelspitze beobachtet: in den isodiametrischen Zellen der meristematischen Zone und in der beginnenden Wurzelhaarzone. Darüber hinaus konnte gezeigt werden, dass die EGFP-Fusionsproteine unter Kontrolle des nativen pMTP2-Promotors nur in epidermalen Zellen exprimiert werden. Für eine homozygote Knockout- Linie, mtp2-S3, konnte bisher kein eindeutiger Phänotyp identifiziert werden. Auf Grundlage der bisher durchgeführten Charakterisierung von AtMTP2 erscheinen zwei Modelle der Funktion von AtMTP2 in der Pflanze möglich: AtMTP2 könnte essentiell für die Versorgung des ER mit Zn unter Zn-Mangelbedingungen sein. Hierfür spricht, dass AtMTP2 in jungen, teilungsaktiven und damit Zn-benötigenden Wurzelzonen exprimiert wird. Die auf die Epidermis beschränkte Lokalisation könnte bei diesem Modell auf die Möglichkeit der zwischenzellulären Zn-Verteilung innerhalb des ER über Desmotubules hindeuten. Alternativ könnte AtMTP2 eine Funktion bei der Detoxifizierung von Zn unter Zn-Schock Bedingungen haben: Es ist bekannt, dass unter Zn- Mangelbedingungen die Expression der zellulären Zn-Aufnahmesysteme hochreguliert wird. Wenn nun die Zn-Verfügbarkeit im Boden z. B durch eine pH-Änderung innerhalb kurzer Zeit stark ansteigt, besteht die Notwendigkeit der Entgiftung von Zn innerhalb der Zelle, bis der starke Einstrom von Zn ins Zytoplasma durch die Deaktivierung der Zn-Aufnahmesysteme und einer geringeren Expression in der Pflanze gedrosselt ist. Ein ähnlicher Mechanismus wurde in der Bäckerhefe S. cerevisae beschrieben, in der darüber hinaus ein Zn-Transporter verstärkt exprimiert wird, der Zn durch Transport in die Vakuole entgiften kann. Es ist durchaus möglich, dass in Arabidopsis AtMTP2 die Zn-Detoxifizierung unter diesen speziellen Bedingungen durch Zn-Transport in das ER oder die Vakuole vermittelt. Zur Identifikation weiterer Komponenten des Metallhomöostasenetzwerks sind verschiedene Ansätze denkbar. In dieser Arbeit wurde in Hefe ein heterologer Screen durchgeführt, um Interaktoren für vier Mitglieder der Arabidopsis-CDF-Familie zu identifizieren. Unter den 11 im Hefesystem bestätigten Kandidaten befindet sich mit AtSPL1 ein AtMTP1-Interaktionskandidat, der möglicherweise eine Rolle bei der Cu-,Zn-Homöostase spielt. Als wahrscheinliche AtMTP3-Interaktionskandidaten wurde die c”-Untereinheit der vakuolären H+-ATPase AtVHA identifiziert sowie mit AtNPSN13 ein Protein, das vermutlich eine Rolle bei Fusionen von Vesikeln mit Zielmembranen spielt. Ein anderer Ansatz zur Identifikation neuer Metallhomöostasegene ist die vergleichende Elementanalyse von natürlichen oder mutagenisierten Pflanzenpopulationen. Voraussetzung für diesen Ansatz ist die schnelle und genaue Analyse des Elementgehalts von Pflanzen. Eine etablierte Methode zur simultanen Bestimmung von bis zu 65 Elementen in einer Probe ist die Inductively Coupled Plasma Optical Emission Spectrometry (ICP OES). Der limitierende Faktor für einen hohen Probendurchsatz ist die Notwendigkeit, Proben für die Analyse zu verflüssigen. Eine alternative Methode der Probenzuführung zum Analysegerät ist die elektrothermale Verdampfung (ETV) der Probe. Zur weitgehend automatisierten Analyse von Pflanzenmaterial mit minimiertem Arbeitsaufwand wurde eine Methode entwickelt, die auf der Kopplung der ETV mit der ICP OES basiert. / All organisms require for their survival essential metals. For each required metal exists an optimal concentration between metal deficiency and -toxicity. It has become evident that all organisms developed a complex network of proteins and low molecular compounds to maintain the equilibrium between all metals. Only few molecular components of this metal-homeostasis network are characterized in detail: A number of protein families whose members transport metals over the barrier of lipid-membranes have been identified during the last couple of years. One of those metal-transport families is the Cation Diffusion Facilitator (CDF) family. All characterized members export metals from the cytoplasm – either into cellular compartments or outside the cell. From the 12 Arabidopsis thaliana (A.thaliana) members – Metal Tolerance Protein (MTP)-1 to 1-2 – only MTP1 and MTP3 have been characterized yet. In this work, characterization of MTP2 is described. As was found for the homologous proteins AtMTP1 and AtMTP3, heterologous expression of AtMTP2 in Zn-sensitive yeast mutants leads to enhanced Zn-tolerance. Less pronounced, enhanced tolerance was also found for Co when AtMTP2 was expressed in Co sensitive yeast mutants. Expression of chimeric AtMTP2/GFP fusion proteins in yeast, A.thaliana protoplasts and in stably transformed A.thalinana plant lines indicated localization of MTP2 in membranes of the endoplasmic reticulum, when GFP was fused to the C-terminal end of MTP2. Fusion of GFP to the N-terminal end of MTP2 lead to vacuolar localization that is most likely explained as mistargeting due to masking of an ER retrieval motive (XXRR) found at the N-terminus of MTP2. This suggests that AtMTP2 mediates the transport of Zn and Co into the endomembrane system of the cell. Tissue specific localization was performed with plant lines expressing the β-Glucuronidase (GUS) reporter protein and with plant lines expressing chimeric fusions of GFP with AtMTP2 under control of the native pMTP2 promoter. Those experiments confirmed Affymetrix Genechip® data suggesting activity of the pMTP2 promoter only under Zn-deficiency. GUS activity was only found under Zn-deficiency in two zone of root tips – the meristematic zone, characterized by isodiamtric cells, and in the beginning differentiation zone, characterized by appearing root hairs. Confocal microscopy with plant lines expressing chimeric MTP2 /GFP fusions demonstrated that expression of AtMTP2 is restricted to epidermal cells. A phenotype for the homozygous mtp2-S3 knockout mutant could not be identified yet. Based on the data obtained as yet / two mode of action of AtMTP2 in planta seam likely: AtMTP2 could be essential for delivery of Zn to the ER under Zn-deficiency. This is supported by the fact, that AtMTP2 is active in young, dividing (and therefore Zn-requiring) zones of the root. The epidermal-restricted expression of AtMTP2 points towards a distribution of Zn in these root zones of Zn within desmotubules. Alternatively, AtMTP2 could have a Zn-detoxifying function under Zn-shock. It is known that in yeast under Zn-deficiency not only the expression of an Zn-uptake transporter is up-regulated, but also the expression of a vacuolar Zn-transporter. It mediates Zn-detoxification of surplus Zn that enters cells upon Zn-resupply before shut down of the Zn uptake system. AtMTP2 could exert this function when soil Zn-availability raises suddenly, for example due to rain after a drought. Different means/methods are perceivable to identify further components of the metal homeostasis network. In this work, a heterologuos screen was performed in yeast to identify interacting proteins for four members of the Arabidopsis CDF-family. Among 11 candidates identified and confirmed in the Split Ubiquitin System (SUS, a Yeast-2-Hybrid variant) is with AtSPL1 an AtMTP1 interaction candidate, which plays putatively a role in Zn,Cu homoestasis. The c” subunit of the vacuolar H+-ATPase AtVHA was found as likely AtMTP3-interaction candidate, as well as AtNPSN13, an protein that plays putatively a role in fusion of vesicles with target-membranes. Another method to identify new metal homeostasis genes is the comparative elemental analysis of natural and mutagenized plant populations. Prerequisite for this approach is the fast and accurate analysis of the elemental composition of plants. An established method for elemental analysis is Inductively Coupled Plasma Optical Emission Spectrometry (ICP OES). The limiting factor for high thoughput is the requirement for laborious wet digest of plant samples before analysis. An alternative mean of sample delivery to the ICP OES is electrothermal vaporization (ETV). For faster, less laborious analysis of plant material, a method based on the established coupling of ETV with ICP OES was developed, which is optimized for plant material and automated as far as possible.

Les métamatériaux à indice de réfraction négative ont attiré énormément l'attention ces dernières années surtout à cause de leurs propriétés électromagnétiques uniques. Ces matériaux sont des structures artificielles qui présentent des caractéristiques n'étant pas disponibles en matériaux naturels. Récemment, le développement technologique avec de nouvelles techniques de fabrication offrent un grand nombre de nouvelles application et développement de nouveaux matériaux. Il est possible d'obtenir un métamatériau en combinant des structures artificielles périodiquement. Les propriétés uniques du Split Ring Resonator (SRR), les Surfaces à Haute Impédance (HIS), les Surface Sélective en Fréquence (FSS) sont étudiées ainsi que les métamatériaux composés. Il a été démontré avec succès l'utilisation pratique de ces structures dans les circuits et les antennes. Il a été confirmé expérimentalement que les métamatériaux pourrait améliorer la performance des structures considérées dans cette thèse, pour des fréquences où la bande interdite électromagnétique se produit.

The aim of this thesis is the development and benchmarking of computational methods for the analysis of high-throughput data from tiling arrays and next-generation sequencing. Tiling arrays have been a mainstay of genome-wide transcriptomics, e.g., in the identification of functional elements in the human genome. Due to limitations of existing methods for the data analysis of this data, a novel statistical approach is presented that identifies expressed segments as significant differences from the background distribution and thus avoids dataset-specific parameters. This method detects differentially expressed segments in biological data with significantly lower false discovery rates and equivalent sensitivities compared to commonly used methods. In addition, it is also clearly superior in the recovery of exon-intron structures. Moreover, the search for local accumulations of expressed segments in tiling array data has led to the identification of very large expressed regions that may constitute a new class of macroRNAs. This thesis proceeds with next-generation sequencing for which various protocols have been devised to study genomic, transcriptomic, and epigenomic features. One of the first crucial steps in most NGS data analyses is the mapping of sequencing reads to a reference genome. This work introduces algorithmic methods to solve the mapping tasks for three major NGS protocols: DNA-seq, RNA-seq, and MethylC-seq. All methods have been thoroughly benchmarked and integrated into the segemehl mapping suite. First, mapping of DNA-seq data is facilitated by the core mapping algorithm of segemehl. Since the initial publication, it has been continuously updated and expanded. Here, extensive and reproducible benchmarks are presented that compare segemehl to state-of-the-art read aligners on various data sets. The results indicate that it is not only more sensitive in finding the optimal alignment with respect to the unit edit distance but also very specific compared to most commonly used alternative read mappers. These advantages are observable for both real and simulated reads, are largely independent of the read length and sequencing technology, but come at the cost of higher running time and memory consumption. Second, the split-read extension of segemehl, presented by Hoffmann, enables the mapping of RNA-seq data, a computationally more difficult form of the mapping task due to the occurrence of splicing. Here, the novel tool lack is presented, which aims to recover missed RNA-seq read alignments using de novo splice junction information. It performs very well in benchmarks and may thus be a beneficial extension to RNA-seq analysis pipelines. Third, a novel method is introduced that facilitates the mapping of bisulfite-treated sequencing data. This protocol is considered the gold standard in genome-wide studies of DNA methylation, one of the major epigenetic modifications in animals and plants. The treatment of DNA with sodium bisulfite selectively converts unmethylated cytosines to uracils, while methylated ones remain unchanged. The bisulfite extension developed here performs seed searches on a collapsed alphabet followed by bisulfite-sensitive dynamic programming alignments. Thus, it is insensitive to bisulfite-related mismatches and does not rely on post-processing, in contrast to other methods. In comparison to state-of-the-art tools, this method achieves significantly higher sensitivities and performs time-competitive in mapping millions of sequencing reads to vertebrate genomes. Remarkably, the increase in sensitivity does not come at the cost of decreased specificity and thus may finally result in a better performance in calling the methylation rate. Lastly, the potential of mapping strategies for de novo genome assemblies is demonstrated with the introduction of a new guided assembly procedure. It incorporates mapping as major component and uses the additional information (e.g., annotation) as guide. With this method, the complete mitochondrial genome of Eulimnogammarus verrucosus has been successfully assembled even though the sequencing library has been heavily dominated by nuclear DNA. In summary, this thesis introduces algorithmic methods that significantly improve the analysis of tiling array, DNA-seq, RNA-seq, and MethylC-seq data, and proposes standards for benchmarking NGS read aligners. Moreover, it presents a new guided assembly procedure that has been successfully applied in the de novo assembly of a crustacean mitogenome. / Diese Arbeit befasst sich mit der Entwicklung und dem Benchmarken von Verfahren zur Analyse von Daten aus Hochdurchsatz-Technologien, wie Tiling Arrays oder Hochdurchsatz-Sequenzierung. Tiling Arrays bildeten lange Zeit die Grundlage für die genomweite Untersuchung des Transkriptoms und kamen beispielsweise bei der Identifizierung funktioneller Elemente im menschlichen Genom zum Einsatz. In dieser Arbeit wird ein neues statistisches Verfahren zur Auswertung von Tiling Array-Daten vorgestellt. Darin werden Segmente als exprimiert klassifiziert, wenn sich deren Signale signifikant von der Hintergrundverteilung unterscheiden. Dadurch werden keine auf den Datensatz abgestimmten Parameterwerte benötigt. Die hier vorgestellte Methode erkennt differentiell exprimierte Segmente in biologischen Daten bei gleicher Sensitivität mit geringerer Falsch-Positiv-Rate im Vergleich zu den derzeit hauptsächlich eingesetzten Verfahren. Zudem ist die Methode bei der Erkennung von Exon-Intron Grenzen präziser. Die Suche nach Anhäufungen exprimierter Segmente hat darüber hinaus zur Entdeckung von sehr langen Regionen geführt, welche möglicherweise eine neue Klasse von macroRNAs darstellen. Nach dem Exkurs zu Tiling Arrays konzentriert sich diese Arbeit nun auf die Hochdurchsatz-Sequenzierung, für die bereits verschiedene Sequenzierungsprotokolle zur Untersuchungen des Genoms, Transkriptoms und Epigenoms etabliert sind. Einer der ersten und entscheidenden Schritte in der Analyse von Sequenzierungsdaten stellt in den meisten Fällen das Mappen dar, bei dem kurze Sequenzen (Reads) auf ein großes Referenzgenom aligniert werden. Die vorliegende Arbeit stellt algorithmische Methoden vor, welche das Mapping-Problem für drei wichtige Sequenzierungsprotokolle (DNA-Seq, RNA-Seq und MethylC-Seq) lösen. Alle Methoden wurden ausführlichen Benchmarks unterzogen und sind in der segemehl-Suite integriert. Als Erstes wird hier der Kern-Algorithmus von segemehl vorgestellt, welcher das Mappen von DNA-Sequenzierungsdaten ermöglicht. Seit der ersten Veröffentlichung wurde dieser kontinuierlich optimiert und erweitert. In dieser Arbeit werden umfangreiche und auf Reproduzierbarkeit bedachte Benchmarks präsentiert, in denen segemehl auf zahlreichen Datensätzen mit bekannten Mapping-Programmen verglichen wird. Die Ergebnisse zeigen, dass segemehl nicht nur sensitiver im Auffinden von optimalen Alignments bezüglich der Editierdistanz sondern auch sehr spezifisch im Vergleich zu anderen Methoden ist. Diese Vorteile sind in realen und simulierten Daten unabhängig von der Sequenzierungstechnologie oder der Länge der Reads erkennbar, gehen aber zu Lasten einer längeren Laufzeit und eines höheren Speicherverbrauchs. Als Zweites wird das Mappen von RNA-Sequenzierungsdaten untersucht, welches bereits von der Split-Read-Erweiterung von segemehl unterstützt wird. Aufgrund von Spleißen ist diese Form des Mapping-Problems rechnerisch aufwendiger. In dieser Arbeit wird das neue Programm lack vorgestellt, welches darauf abzielt, fehlende Read-Alignments mit Hilfe von de novo Spleiß-Information zu finden. Es erzielt hervorragende Ergebnisse und stellt somit eine sinnvolle Ergänzung zu Analyse-Pipelines für RNA-Sequenzierungsdaten dar. Als Drittes wird eine neue Methode zum Mappen von Bisulfit-behandelte Sequenzierungsdaten vorgestellt. Dieses Protokoll gilt als Goldstandard in der genomweiten Untersuchung der DNA-Methylierung, einer der wichtigsten epigenetischen Modifikationen in Tieren und Pflanzen. Dabei wird die DNA vor der Sequenzierung mit Natriumbisulfit behandelt, welches selektiv nicht methylierte Cytosine zu Uracilen konvertiert, während Methylcytosine davon unberührt bleiben. Die hier vorgestellte Bisulfit-Erweiterung führt die Seed-Suche auf einem reduziertem Alphabet durch und verifiziert die erhaltenen Treffer mit einem auf dynamischer Programmierung basierenden Bisulfit-sensitiven Alignment-Algorithmus. Das verwendete Verfahren ist somit unempfindlich gegenüber Bisulfit-Konvertierungen und erfordert im Gegensatz zu anderen Verfahren keine weitere Nachverarbeitung. Im Vergleich zu aktuell eingesetzten Programmen ist die Methode sensitiver und benötigt eine vergleichbare Laufzeit beim Mappen von Millionen von Reads auf große Genome. Bemerkenswerterweise wird die erhöhte Sensitivität bei gleichbleibend guter Spezifizität erreicht. Dadurch könnte diese Methode somit auch bessere Ergebnisse bei der präzisen Bestimmung der Methylierungsraten erreichen. Schließlich wird noch das Potential von Mapping-Strategien für Assemblierungen mit der Einführung eines neuen, Kristallisation-genanntes Verfahren zur unterstützten Assemblierung aufgezeigt. Es enthält Mapping als Hauptbestandteil und nutzt Zusatzinformation (z.B. Annotationen) als Unterstützung. Dieses Verfahren ermöglichte die erfolgreiche Assemblierung des kompletten mitochondrialen Genoms von Eulimnogammarus verrucosus trotz einer vorwiegend aus nukleärer DNA bestehenden genomischen Bibliothek. Zusammenfassend stellt diese Arbeit algorithmische Methoden vor, welche die Analysen von Tiling Array, DNA-Seq, RNA-Seq und MethylC-Seq Daten signifikant verbessern. Es werden zudem Standards für den Vergleich von Programmen zum Mappen von Daten der Hochdurchsatz-Sequenzierung vorgeschlagen. Darüber hinaus wird ein neues Verfahren zur unterstützten Genom-Assemblierung vorgestellt, welches erfolgreich bei der de novo-Assemblierung eines mitochondrialen Krustentier-Genoms eingesetzt wurde.

Submitted by Milena Rubi (milenarubi@ufscar.br) on 2017-10-09T16:26:41Z No. of bitstreams: 1 CRUZ_Jadder_2017.pdf: 1326879 bytes, checksum: 61ee3c40e293d26085a939c0a0290716 (MD5) / Approved for entry into archive by Milena Rubi (milenarubi@ufscar.br) on 2017-10-09T16:26:55Z (GMT) No. of bitstreams: 1 CRUZ_Jadder_2017.pdf: 1326879 bytes, checksum: 61ee3c40e293d26085a939c0a0290716 (MD5) / Approved for entry into archive by Milena Rubi (milenarubi@ufscar.br) on 2017-10-09T16:27:03Z (GMT) No. of bitstreams: 1 CRUZ_Jadder_2017.pdf: 1326879 bytes, checksum: 61ee3c40e293d26085a939c0a0290716 (MD5) / Made available in DSpace on 2017-10-09T16:27:11Z (GMT). No. of bitstreams: 1 CRUZ_Jadder_2017.pdf: 1326879 bytes, checksum: 61ee3c40e293d26085a939c0a0290716 (MD5) Previous issue date: 2017-05-02 / Coordenação de Aperfeiçoamento de Pessoal de Nível Superior (CAPES) / Let G = (V, E) be a simple and undirected graph. An edge-coloring is an assignment of colors to the edges of the graph such that any two adjacent edges receive different colors. The chromatic index of a graph G is the smallest number of colors such that G has an edge-coloring. Clearly, a lower bound for the chromatic index is the degree of the vertex of higher degree, denoted by ?(G). In 1964, Vizing proved that chromatic index is ?(G) or ?(G) + 1. The Classification Problem is to determine if the chromatic index is ?(G) (Class 1 ) or if it is ?(G) + 1 (Class 2 ). Let n be number of vertices of a graph G and let m be its number of edges. We say G is overfull if m > (n-1) 2 ?(G). Every overfull graph is Class 2. A graph is subgraph-overfull if it has a subgraph with same maximum degree and it is overfull. It is well-known that every overfull and subgraph-overfull graph is Class 2. The Overfull Conjecture asserts that every graph with ?(G) > n 3 is Class 2 if and only if it is subgraph-overfull. In this work we prove the Overfull Conjecture to a particular class of graphs, known as split-comparability graphs. The Overfull Conjecture was open to this class. / Dado um grafo simples e não direcionado G = (V, E), uma coloração de arestas é uma função que atribui cores às arestas do grafo tal que todas as arestas que incidem em um mesmo vértice têm cores distintas. O índice cromático é o número mínimo de cores para obter uma coloração própria das arestas de um grafo. Um limite inferior para o índice cromático é, claramente, o grau do vértice de maior grau, denotado por ?(G). Em 1964, Vizing provou que o índice cromático ou é ?(G) ou ?(G) + 1, surgindo assim o Problema da Classificação, que consiste em determinar se o índice cromático é ?(G) (Classe 1 ) ou ?(G) + 1 (Classe 2 ). Seja n o número de vértices de um grafo G e m seu número de arestas. Dizemos que um grafo é sobrecarregado se m > (n-1) 2 ?(G). Um grafo é subgrafo-sobrecarregado se tem um subgrafo de mesmo grau máximo que é sobrecarregado. É sabido que se um grafo é sobrecarregado ou subgrafo-sobrecarregado ele é necessariamente Classe 2. A Conjectura Overfull é uma famosa conjectura de coloração de arestas e diz que um grafo com ?(G) > n 3 é Classe 2 se e somente se é subgrafo-sobrecarregado. Neste trabalho provamos a Conjectura Overfull para uma classe de grafos, a classe dos grafos split-comparabilidade. Até este momento a Conjectura Overfull estava aberta para esta classe.

Made available in DSpace on 2015-03-26T13:51:52Z (GMT). No. of bitstreams: 1 texto completo.pdf: 800491 bytes, checksum: 274cc18672a32ed45926e416e0de5305 (MD5) Previous issue date: 2011-02-15 / The homologue recombination in Kluyveromyces lactis is not the preferential way used as repair mechanism of DNA double strand, desirable in proteins expression construction dependent on gene-specific integration. In order to obtain K. lactis strains to recombinant protein expression efficient in homologue recombination way, KU80 gene was interrupted. The perfect running of non-homologue ends junctions (NHEJ) way depends on that gene and is responsible by exogenous DNA random integration in the host genome. KU80 gene deletion was made by Split-Marker. Two fragments were generated by PCR fusion, each one with a flanker sequence of KU80 coding region (5 and 3 ends) and part of geneticin resistance gene (KanMX). Deletion cassette resulting from in vivo recombination of two fragments had KanMX gene flanked by KU80 coding regions end sand was used for KU80 deletion by homologue recombination in the genome in two K. lactis strains, JA6 and HP108. The 3.7 Kb fragment obtained by PCR amplification with external primers to KU80 coding region confirmed deletion cassette integration in the target gene. Integration efficiency with Split-Marker resulting fragments was 100 %. pKLAC1 and pKLAC1/cStp vectors with streptavidin affinity domain by biotin were used to determine homologue recombination efficiency of KU80 (JA6ΔKU80 and HP108ΔKU80) mutant strains. Pectin lyase coding gene (plg1) from Penicillium griseoroseum was cloned in those vectors to evaluate the capacity of K. lactis strains to produce and secrete the recombinant protein. The transformation efficiency (transformants/μg of DNA) of mutants JA6ΔKU80 and HP108ΔKU80 with pKLAC1/Plg1 and pKLAC1/cStp-Plg1 vectors was higher than to parental strains JA6 and HP108. Target gene integration efficiency was 100 % in most strains, except to strains JA6/Plg1and HP108ΔKU80/Plg1 that showed an integration efficiency of 80 and 70 %, respectively. Although the high efficiency of specific-gene integration there was no pectin lyase (PL) or cStp-Plg1 secretion as expected for pKLAC1 vector. PL intracellular activity was significant when compared with parental strain HP108/Plg1, that presented specific activity of 9,525 U.mg-1 protein. / A recombinação homóloga em Kluyveromyces lactis não é a via preferencial usada como mecanismo de reparo de quebra de fita dupla de DNA, o que pode ser indesejado em construções de expressão de proteínas dependentes de integração gene-específico. Para obter linhagens de K. lactis hospedeiras para a expressão de proteínas recombinantes eficientes na via de recombinação homóloga realizou-se a mutação do gene KU80. Este gene é essencial para perfeito funcionamento da via de junções de extremidades não-homólogas (NHEJ), responsável pela integração aleatória do DNA exógeno no genoma hospedeiro. A deleção do gene KU80 foi feita utilizando a técnica Split-Marker. Dois fragmentos foram obtidos por fusão por PCR, cada um contendo uma sequência flanqueadora da região codificante do gene KU80 (extremidades 5 e 3 ) e uma parte do gene de resistência a geneticina (KanMX). O cassete de deleção resultante da recombinação in vivo dos dois fragmentos gerados continha o gene KanMX flanqueado pelas extremidades da região codificante do gene KU80, e foi utilizado para deleção do gene KU80 por recombinação homóloga no genoma de duas linhagens de K. lactis, JA6 e HP108. O fragmento de 3,7 Kb obtido por amplificação por PCR com oligonucleotídeos externos à região codificante do gene KU80 confirmou a integração do cassete de deleção no gene alvo. A eficiência de integração com os fragmentos resultantes do Split-Marker foi de 100 %. Os vetores pKLAC1, e pKLAC1/cStp contendo o domínio de afinidade da estreptavidina pela biotina, foram usados para determinar a eficiência de recombinação homóloga das linhagens mutantes KU80 (JA6ΔKU80 e HP108ΔKU80). O gene da pectina liase (plg1) de Penicillium griseoroseum foi clonado nesses vetores para avaliar a capacidade de linhagens de K. lactis em produzir e secretar a proteína recombinante. A eficiência de transformação (transformantes/μg de DNA) dos mutantes JA6ΔKU80 e HP108ΔKU80 com os vetores pKLAC1/Plg1 e pKLAC1/cStp-Plg1, foi superior à das linhagens parentais JA6 e HP108. A eficiência de integração no gene alvo foi de 100% para maioria das linhagens, com exceção das linhagens JA6/Plg1e HP108ΔKU80/Plg1 que apresentaram, respectivamente, eficiência de 80 e 70 % de integração geneespecífico. Apesar da eficiência de integração por recombinação homóloga, não houve secreção de PL e da fusão cStp-Plg1 como esperado ao se utilizar o vetor pKLAC1. A atividade intracelular de pectina liase (PL) só foi significativa em relação à parental para a cepa HP108/Plg1, que demonstrou atividade específica de 9,525 U.mg-1 proteína.

Made available in DSpace on 2015-03-26T13:51:58Z (GMT). No. of bitstreams: 1 texto completo.pdf: 952076 bytes, checksum: 7fbcb43414ecacd3439620826b6172cd (MD5) Previous issue date: 2012-07-20 / Conselho Nacional de Desenvolvimento Científico e Tecnológico / Colletotrichum lindemuthianum is the causal agent of common bean anthracnose. Genes that encode cell wall-degrading enzymes are essential for the development of this disease. The pectinases are characterized as the most important group of cell wall- degrading enzymes produced by phytopathogen fungi. The gene coding for pectate lyase, pecCl1, was previously identified in a suppressive subtractive library of bean infected with C. lindemuthianum. Isolation of the gene pecCl1 made it possible to obtain mutants and to analyze the regulation of this gene during development of anthracnose, determining whether the pectate lyase is a pathogenic factor. Thus, the aim of our study was structurally and functionally characterize the gene encoding pectate lyase in C. lindemuthianum. Initially, was performed the structural analysis of the gene pecCl1. The complete nucleotide sequence of the gene pecCl1 was deposited in Genbank with accession number JX270683. The analysis of the promoter region revealed some putative cis-elements and potential binding motifs of transcription factors involved in the regulation of pectate lyase gene expression. The deduced amino acid sequence of pecCl1 showed sequence identity with the pectate lyase F of Colletotrichum higginsianum and the pectate lyase C of Glomerella graminicola M1.001. Furthermore, it was found putative conserved domain pfam03211 of the pectate lyases superfamily. The gene pecCl1 is represented by a single copy in the C. lindemuthianum genome. However, into the genome of Colletotrichum graminicola, three sequences encoding pectate lyase showed sequence identity with the gene pecCl1 of C. lindemuthianum, and into the genome of C. higginsianum seven sequences encoding pectate lyase showed sequence identity with the gene pecCl1 of C. lindemuthianum, indicating that the C. lindemuthianum genome can possess other genes encoding pectate lyase. Phylogenetic analysis of pectate lyase amino acid sequences of filamentous fungi exhibited the formation of two distinct groups which are grouped on the basis of members of the pectate lyases multigene family. The Split-Marker technique was effective in C. lindemuthianum pecCl1 gene inactivation, allowing the study of pecCl1 function in a mutant by specific integrations and without ectopic integrations. The pecCl1 gene inactivation did not lead to complete loss of the pectate lyase activity, and consequently only decreased anthracnose symptoms in its host, which is consistent with the presence of other genes coding pectate lyase, allowing greater flexibility in pathogen aggressiveness. The analysis of differential expression of gene pecCl1 by qPCR was performed at different stages of bean infection and were observed expression levels of pecCl1 at all stages of development of the fungus in the plant, but a significant increase was observed five days after infection, in the onset of necrotrophic stage. At this stage, secondary hyphae cause extensive degradation of plant cell wall through the secretion of wide range of depolymerases, among these, the pectate lyase. Thus, the pectate lyase encoded by the gene pecCl1 is important to aggressiveness of C. lindemuthianum. The analysis of pectate lyases in C. lindemuthianum can not only assist in understanding the disease, but may also lead to discovery of one more target for disease control. / Colletotrichum lindemuthianum é o agente causal da antracnose do feijoeiro comum. Genes que codificam enzimas que degradam a parede celular são essenciais para o desenvolvimento dessa doença. As pectinases são caracterizadas como o grupo de enzimas que hidrolisam a parede celular mais importante produzidas por fungos fitopatogênicos. O gene pecCl1, que codifica pectato liase, foi previamente identificado em uma biblioteca subtrativa supressiva de feijoeiro infectado com C. lindemuthianum. O isolamento do gene tornou possível a obtenção de mutantes e análise da regulação deste gene durante o desenvolvimento da antracnose, visando determinar se a pectato liase é um fator de patogenicidade. Desta forma, o objetivo do nosso trabalho foi caracterizar estruturalmente e funcionalmente o gene que codifica pectato liase em C. lindemuthianum. Inicialmente, foi realizada a análise estrutural do gene pecCl1. A sequência completa de nucleotídeos do gene pecCl1 foi deposita no Genbank com número de acesso JX270683. A análise da região promotora revelou alguns possíveis cis-elementos e sítios de ligação a fatores de transcrição envolvidos na regulação da expressão gênica da pectato liase. A sequência de aminoácidos deduzida de pecCl1 apresentou identidade de sequências com a pectato liase F de Colletotrichum higginsianum e a pectato liase C de Glomerella graminicola M1.001. Além disso, detectou-se um possível domínio conservado pfam03211 da superfamília de pectato liases. O gene pecCl1 encontra-se representado por uma cópia única no genoma de C. lindemuthianum. No entanto, no genoma de Colletotrichum graminicola, três sequências que codificam pectato liase apresentaram identidade de sequências com o gene pecCl1 de C. lindemuthianum, e no genoma de C. higginsianum sete sequências que codificam pectato liase apresentaram identidade de sequências com o gene pecCl1 de C. lindemuthianum, indicando que o genoma de C. lindemuthianum pode possuir além do gene pecCl1 outros genes que codificam pectato liase. A análise filogenética de sequências de aminoácidos de pectato liases de fungos filamentosos mostrou a formação de dois grupos distintos, que se agruparam com base nos membros da família multigênica de pectato liases. A técnica de Split-Marker mostrou-se eficiente na inativação do gene pecCl1 de C. lindemuthianum, possibilitando o estudo da função do gene pecCl1, em um mutante com integração específica e livre de integrações ectópicas. A inativação do gene pecCl1 não levou a perda completa da atividade de pectato liase, e consequentemente, somente diminuiu os sintomas de antracnose em seu hospedeiro, o que é consistente com a presença de outros genes que codificam pectato liase no fungo, permitindo ao patógeno uma maior flexibilidade em sua agressividade. Foi realizada a análise da expressão diferencial do gene pecCl1 por qPCR nos diferentes estágios de infecção no feijoeiro e foram observados transcritos de pecCl1 em todas as fases de desenvolvimento do fungo na planta, mas houve um aumento significativo destes transcritos cinco dias após a infecção, no início da fase necrotrófica do fungo. Nesta fase, as hifas secundárias causam degradação extensiva da parede celular vegetal por meio da secreção de vasta gama de despolimerases, dentre estas, a pectato liase. Portanto, a pectato liase codificada pelo gene pecCl1 é importante para agressividade de C. lindemuthianum. A análise de pectato liases poderá não somente auxiliar na compreensão da antracnose em feijoeiro comum, mas também poderá levar a descoberta de mais um alvo para o controle dessa doença.