Background: Phosphorylation of proteins plays a crucial role in the regulation and activation of metabolic and signaling pathways and constitutes an important target for pharmaceutical intervention. Central to the phosphorylation process is the recognition of specific target sites by protein kinases followed by the covalent attachment of phosphate groups to the amino acids serine, threonine, or tyrosine. The experimental identification as well as computational prediction of phosphorylation sites (P-sites) has proved to be a challenging problem. Computational methods have focused primarily on extracting predictive features from the local, one-dimensional sequence information surrounding phosphorylation sites. Results: We characterized the spatial context of phosphorylation sites and assessed its usability for improved phosphorylation site predictions. We identified 750 non-redundant, experimentally verified sites with three-dimensional (3D) structural information available in the protein data bank (PDB) and grouped them according to their respective kinase family. We studied the spatial distribution of amino acids around phosphorserines, phosphothreonines, and phosphotyrosines to extract signature 3D-profiles. Characteristic spatial distributions of amino acid residue types around phosphorylation sites were indeed discernable, especially when kinase-family-specific target sites were analyzed. To test the added value of using spatial information for the computational prediction of phosphorylation sites, Support Vector Machines were applied using both sequence as well as structural information. When compared to sequence-only based prediction methods, a small but consistent performance improvement was obtained when the prediction was informed by 3D-context information. Conclusion: While local one-dimensional amino acid sequence information was observed to harbor most of the discriminatory power, spatial context information was identified as relevant for the recognition of kinases and their cognate target sites and can be used for an improved prediction of phosphorylation sites. A web-based service (Phos3D) implementing the developed structurebased P-site prediction method has been made available at http://phos3d.mpimp-golm.mpg.de.

Après mes travaux de thèse effectués au Laboratoire l'Electricité, Signaux, et Robotique, (LESiR) de l'Ecole Normale Supérieure de Cachan, devenu depuis Laboratoire des Systèmes et Applications des Technologies de l'Information et de l'Energie (SATIE), j'ai eu la chance de travailler au sein de différentes équipes de recherche en génie électrique, tout d'abord, en Grande Bretagne, à l'"Electrical Machines and Drives Group" de l'Université de Sheffield, puis à Belfort au Laboratoire d'Electronique, Electrotechnique et Systèmes (L2ES), et enfin au Havre, au sein du Groupe de Recherche en Electrotechnique et Automatique du Havre (GREAH). Ce document me permet de présenter ces travaux afin de solliciter une Habilitation à Diriger des Recherches. Ce mémoire comporte trois parties. Une première partie où sont regroupés mon CV, une présentation de mes activités d'enseignement, d'administration, d'encadrement et de recherche et enfin une énumération des publications auxquelles j'ai participé. Dans la seconde partie, qui constitue le cœur de ce mémoire, une présentation plus approfondie de mes thématiques de recherche sera abordée. Ces thématiques sont regroupées en deux catégories : l'une plus fondamentale, du point de vue des sciences de l'ingénieur, concerne la modélisation des machines électriques, et la seconde plus applicative concerne les machines synchrones à double excitation. Alors que lors de mes travaux de thèse, qui concernaient les machines synchrones à double excitation, j'ai été plus utilisateur que développeur de méthodes de modélisation, je me suis de plus en plus orienté vers le développement de modèles au cours de mon séjour à l'Université de Sheffield et maintenant au GREAH. Les modèles développés sont bien adaptés pour le dimensionnement optimal des machines électriques. Ils sont de deux natures : * modèles analytiques ; * modèles semi-numériques. Les modèles analytiques sont basés sur la résolution formelle des équations de Maxwell dans les régions à perméabilité uniforme. Par modèles semi-numériques on entend les modèles basés sur les réseaux de permeances. La thématique des machines synchrones à double excitation, bien que n'étant pas une problématique fondamentale à proprement dit, permet d'apporter des réponses à des problématiques qui ne le sont pas moins : "défluxage" des machines à aimants permanents, optimisation énergétique, et enfin augmentation du coût des aimants permanents. Cette thématique est essentiellement menée en collaboration avec des collègues de l'équipe Systèmes d'Énergies pour les Transports et l'Environnement (SETE) du SATIE. Enfin, dans la troisième partie un bilan global des activités de recherche est dressé. Ce bilan permettra de revenir sur les points que je considère essentiels de ces activités, et de dégager quelques perspectives aux court, moyen, et long termes.

Le développement de machines à forte vitesse spécifique et de taille réduite en régime subsonique suscite actuellement une forte demande dans de nombreux domaines industriels. Les machins à rotors contrarotatifs largement étudiées en aéronautique constituent une alternative efficace aux machines conventionnelles offrant de nombreux avantages : réduction de la vitesse de rotation, de l'encombrement radial et une grande flexibilité d'utilisation. Cependant, leur utilisation dans des applications subsoniques courantes nécessite une meilleure compréhension de leur fonctionnement et notamment de l'interaction inter-rotors pour mieux les concevoir. Ce travail a pour objectif d'étudier et de caractériser expérimentalement un étage contrarotatif fonctionnant en conduit, conçu avec le code de conception et d'analyse pour rotor et rotor-stator, MFT auquel on a implémenté une méthode de conception simple et rapide pour dessiner le rotor aval. On analyse en particulier l'effet de la distance axiale entre les rotors et l'effet du rapport de leur vitesse. Il met en évidence une nette amélioration des caractéristiques et du rendement global par rapport à une machine conventionnelle. Par ailleurs, plusieurs aspects de l'interaction entre les rotors sont constatés à travers des mesures locales à proximité des rotors. Le présent mémoire s'articule autour de quatre parties : conception d'un étage rotor-stator et d'un étage contrarotatif ; conception du dispositif expérimental normalisé, AERO²FANS pour les mesures de performances globales et locales instationnaires ; caractérisation et comparaison des deux étages et validation de la conception de MFT; enfin, étude des effets de la distance axiale et du rapport des vitesses sur les performances globales et locales.

Communication complexity is an area of complexity theory that studies an abstract model of computation called a communication protocol. In a $k$-player communication protocol, an input to a known function is partitioned into $k$ pieces of $n$ bits each, and each piece is assigned to one of the players in the protocol. The goal of the players is to evaluate the function on the distributed input by using as little communication as possible. In a Number-On-Forehead (NOF) protocol, the input piece assigned to each player is metaphorically placed on that player's forehead, so that each player sees everyone else's input but its own. In a Number-In-Hand (NIH) protocol, the piece assigned to each player is seen only by that player. Overall, the study of communication protocols has been used to obtain lower bounds and impossibility results for a wide variety of other models of computation. Two of the main contributions presented in this thesis are negative results on the NOF model of communication, identifying limitations of NOF protocols. Together, these results consitute stepping stones towards a better fundamental understanding of this model. As the first contribution, we show that randomized NOF protocols are exponentially more powerful than deterministic NOF protocols, as long as $k \le n^c$ for some constant $c$. As the second contribution, we show that nondeterministic NOF protocols are exponentially more powerful than randomized NOF protocols, as long as $k \le \delta \cdot \log n$ for some constant $\delta < 1$. For the third major contribution, we turn to the NIH model and we present a positive result. Informally, we show that a NIH communication protocol for a function $f$ can simulate a Stack Machine (a Turing Machine augmented with a stack) for a related function $F$, consisting of several instances of $f$ bundled together. Using this simulation and known communication complexity lower bounds, we obtain the first known (space vs. number of passes) trade-off lower bounds for Stack Machines.

Recently, many methods have been proposed for the classification and prediction problems in bioinformatics. One of these problems is the protein structure prediction. Machine learning approaches and new algorithms have been proposed to solve this problem. Among the machine learning approaches, Support Vector Machines (SVM) have attracted a lot of attention due to their high prediction accuracy. Since protein data consists of sequence and structural information, another most widely used approach for modeling this structured data is to use graphs. In computer science, graph theory has been widely studied; however it has only been recently applied to bioinformatics. In this work, we introduced new algorithms based on statistical methods, graph theory concepts and machine learning for the protein structure prediction problem. A new statistical method based on z-scores has been introduced for seed selection in proteins. A new method based on finding common cliques in protein data for feature selection is also introduced, which reduces noise in the data. We also introduced new binary classifiers for the prediction of structural transitions in proteins. These new binary classifiers achieve much higher accuracy results than the current traditional binary classifiers.

Bacterial cells were once treated as membrane-enclosed bags of cytoplasm: a homogeneous, undifferentiated suspension in which polymers (proteins, nucleic acids, etc.) and small molecules diffused freely to interact with each other. Biochemical studies have determined the molecular mechanisms underlying the biological processes of metabolism, replication and transcription-translation, etc. However, recent advancements in optical techniques armed with fluorescent tags for proteins and nucleic acids have increased our ability to peer into the interior of live bacterial cells. This has revealed an organized layout of multi-protein complexes, or molecular machines, dedicated to specific functions at defined sub-cellular locations; the timing of their assembly and/or rates of their activity being determined by available nutrition and environmental signals from the niche occupied by the organism. In the present study, we have attempted to identify the intracellular location and organization of the molecular machines assembled for protein synthesis (ribosomes), DNA replication (replisomes) and cell division (divisome) in different bacteria. We have used the model system Escherichia coli as well as Helicobacter pylori and mycobacterial strains (Mycobacterium marinum and Mycobacterium smegmatis), which grow at different rates and move to dormancy late into stationary phase Bacterial nucleoid plays a major role in organizing the location and movement of active ribosomes, replisomes and placement of divisome. While the active ribosomes appear to follow the dynamic folds of the bacterial nucleoid during cell growth in E. coli, inactive ribosomes appear to accumulate near the periphery. The replisome in H. pylori was visualized as a sharp, single focus upon SSB and DnaB co-localization in growing helical rods but disassembled into diffused fluorescence when the cells attained non-replicative coccoid stage. Our investigation into mycobacterial life-cycle revealed unique features such as an absence of a dedicated mid-cell site for divisome assembly and endosporulation upon entry into stationary phase. In brief, we present the cell cycle-dependent subcellular organization of molecular machines in bacteria.

The problem of identifying genes in eukaryotic genomic sequences by computational methods has attracted considerable research attention in recent years. Many early approaches to the problem focused on prediction of individual functional elements and compositional properties of coding and non coding deoxyribonucleic acid (DNA) in entire eukaryotic gene structures. More recently, a number of approaches has been developed which integrate multiple types of information including structure, function and genetic properties of proteins. Knowledge of the structure of a protein is essential for describing and understanding its function. In addition, subcellular localization of a protein can be used to provide some amount of characterization of a protein. In this study, a method for the prediction of protein subcellular localization based on primary sequence data is described. Primary sequence data for a protein is based on amino acid sequence. The frequency value for each amino acid is computed in one given position. Assigned frequencies are used in a new encoding scheme that conserves biological information based on point accepted mutations (PAM) substitution matrix. This method can be used to predict the nuclear, the cytosolic sequences, the mitochondrial targeting peptides (mTP) and the signal peptides (SP). For clustering purposes, other than well known traditional techniques, principle component analysis (PCA)&quot / and self-organizing maps (SOM)&quot / are used. For classication purposes, support vector machines (SVM)&quot / , a method of statistical learning theory recently introduced to bioinformatics is used. The aim of the combination of feature extraction, clustering and classification methods is to design an acccurate system that predicts the subcellular localization of proteins presented into the system. Our scheme for combining several methods is cascading or serial combination according to its architecture. In the cascading architecture, the output of a method serves as the input of the other model used.

The thesis reports the results of comprehensive studies carried out to explore the impact of midpoint FACTS Controllers (STATCOM and SVC) on the generator distance phase backup protection in order to identify important issues that protection engineers need to consider when designing and setting a generator protection system. In addition, practical, feasible and simple solutions to mitigate the adverse impact of midpoint FACTS Controllers on the generator distance phase backup protection are explored. The results of these studies show that midpoint FACTS Controllers have an adverse effect on the generator distance phase backup protection. This adverse effect, which can be in the form of underreach, overreach or a time delay, varies according to the fault type, fault location and generator loading. Moreover, it has been found that the adverse effect of the midpoint FACTS Controllers extends to affect the coordination between the generator distance phase backup protection and the generator steady-state overexcited capability limit. The Support Vector Machines classification technique is proposed as a replacement for the existing generator distance phase backup protection relay in order to alleviate potential problems. It has been demonstrated that this technique is a very promising solution, as it is fast, reliable and has a high performance efficiency. This will result in enhancing the coordination between the generator phase backup protection and the generator steady-state overexcited capability limit in the presence of midpoint FACTS Controllers. The thesis also presents the results of investigations carried out to explore the impact of the generator distance phase backup protection relay on the generator overexcitation thermal capability. The results of these investigations reveal that with the relay settings according to the current standards, the generator is over-protected and the generator distance phase backup protection relay restricts the generator overexcitation thermal capability during system disturbances. This restriction does not allow the supply of the maximum reactive power of the generating unit during such events. The restriction on the generator overexcitation thermal capability caused by the generator distance phase backup protection relay highlights the necessity to revise the relay settings. The proposed solution in this thesis is to reduce the generator distance phase backup protection relay reach in order to provide secure performance during system disturbances.

Communication complexity is an area of complexity theory that studies an abstract model of computation called a communication protocol. In a $k$-player communication protocol, an input to a known function is partitioned into $k$ pieces of $n$ bits each, and each piece is assigned to one of the players in the protocol. The goal of the players is to evaluate the function on the distributed input by using as little communication as possible. In a Number-On-Forehead (NOF) protocol, the input piece assigned to each player is metaphorically placed on that player's forehead, so that each player sees everyone else's input but its own. In a Number-In-Hand (NIH) protocol, the piece assigned to each player is seen only by that player. Overall, the study of communication protocols has been used to obtain lower bounds and impossibility results for a wide variety of other models of computation. Two of the main contributions presented in this thesis are negative results on the NOF model of communication, identifying limitations of NOF protocols. Together, these results consitute stepping stones towards a better fundamental understanding of this model. As the first contribution, we show that randomized NOF protocols are exponentially more powerful than deterministic NOF protocols, as long as $k \le n^c$ for some constant $c$. As the second contribution, we show that nondeterministic NOF protocols are exponentially more powerful than randomized NOF protocols, as long as $k \le \delta \cdot \log n$ for some constant $\delta < 1$. For the third major contribution, we turn to the NIH model and we present a positive result. Informally, we show that a NIH communication protocol for a function $f$ can simulate a Stack Machine (a Turing Machine augmented with a stack) for a related function $F$, consisting of several instances of $f$ bundled together. Using this simulation and known communication complexity lower bounds, we obtain the first known (space vs. number of passes) trade-off lower bounds for Stack Machines.

The proliferation of electric machines has drastically increased in recent years and is likely to increase into the future. This interest in the production of advanced, high power density electrical machines that are small in size has heightened the need for flexible manufacturing processes to produce their laminated components during short batch and prototyping production runs. A means of cost effective, accelerated prototyping of these machines will have a substantial impact on their design and optimization, reducing time requirements to produce and test a given design. A review of the current manufacturing methods for prototyping electric machines was conducted. In particular, laser cutting, electric discharge machining, and abrasive waterjet (AWJ) machining were researched as competitive processes. Each of these methods exhibits marked advantages and disadvantages that present the opportunity for a new process to compete. This work investigates the applicability of high throughput profile milling (HTPM) for the prototyping of advanced electric machines, specifically, the process parameter space for milling of electrical steels. The material response will be determined by characterizing its specific cutting energy and utilizing this to develop a model to predict cutting forces during the milling process. Optimal process parameters will be investigated to obtain maximum productivity and minimal burr formation. Finally, the impact of HTPM processing on the magnetic properties of electrical steels will be compared to that of a leading prototyping technology, AWJ machining.