Protein aggregation is commonly encountered during the manufacture of protein-based bioproducts in processing such as protein expression, purification, refolding, shipping and storage (Volkin and Middaugh, 1992; Brange, 2000). Aggregation may shorten the shelf-life of pharmaceutical proteins (Frokjaer and Otzen, 2005) and induce severe hypersensitivity (Rosenberg, 2006). In addition, several diseases ranging from Alzheimer’s disease to cystic fibrosis are associated with protein aggregation in the form of amyloid fibrils and plaques (Dobson, 1999; Luheshi et al., 2008). Hence, studies on protein aggregation, especially those dealing with high concentrations of proteins, are highly demanded in both academic and industrial laboratories. To address the aforementioned issues, physicochemical factors affecting protein aggregation were investigated systematically in this project. Strategies were developed to inhibit protein aggregation during renaturation and to enhance protein stability against aggregation during and after production, especially when dealing with high protein concentrations. ∆5-3-Ketosteroid isomerase (KSI) was used as a model for aggregation studies during protein renaturation due to its intrinsic aggregation properties. KSI was overexpressed as inclusion bodies (IBs) in Escherichia coli (E. coli). Cost- and time-efficient combination of chemical extraction and one-step affinity purification ensured the production of denatured KSI with high purity at high yield. Several key factors, including protein concentration and ionic strength, were determined to greatly influence KSI aggregation during renaturation. Polymer addition (PEG 3000 and Eudragit S-100) was found to alter KSI aggregation behaviour in a polymer-specific manner, as quantified using reversed phase-high performance liquid chromatography (RP-HPLC) analysis. Light scattering for second virial coefficient (SVC) measurement, surface plasmon resonance (SPR), and microfluidics were applied to study the fundamental mechanism of protein aggregation. Lysozyme was further introduced as a control protein for comparison with KSI. A rapid lumped method was established to measure specific refractive index (∂n/∂c) and SVC values for KSI and lysozyme, which provided quantitative and qualitative information on thermodynamic interactions of molecules in solution. SPR and microfluidics were also used to explore protein aggregation properties. To our best knowledge, it is the first time SPR and microfluidics have been used to investigate protein aggregation behaviour. Both SPR and microfluidics present significant potential for assessing protein aggregation and diagnosis or drug screening of protein aggregation related diseases. The chemical and physical stability of proteins needs to be maintained after successful refolding to ensure an acceptably long shelf life, especially at high protein concentration (Chang and Hermsdorf, 2002). The pharmaceutical effects of lectins on cell growth provided incentive for studies to improve their stability. Human galectin-2 (hGal-2, a homodimeric lectin) was used as a study model in this project. Mutations were introduced at one of the two Cys residues (C57A, C57M, and C57S). Only the C57M variant was highly expressed in bacteria in soluble form. No aggregate of this mutant was detected during 3 weeks of storage. hGal-2 C57M also facilitated site-directed introduction of poly(ethylene glycol) (PEG) into the remaining sulfhydryl group (Cys75). Product analysis revealed rather complete conjugation with one PEG chain per protein subunit in homodimer. Neither secondary structure alteration nor the absence of binding ability to a glycoprotein (asialofetuin) was observed. The results document the feasibility of tailoring a human galectin for enhanced stability against aggregation as well as monoPEGylation, which enables further testing of biological properties including functionality as a growth regulator and the serum clearance rate of hGal-2.

Universiẗat, Diss., 2003--Heidelberg. / Enth.: 3 Sonderabdr. aus verschiedenen Publikationen. Beitr. teilw. dt., teilw. engl.

Thesis (Ph. D.)--Miami University, Dept. of Geology, 2007. / Title from second page of PDF document. Includes bibliographical references.

Η συγκόλληση των αιμοπεταλίων προάγεται από το ινωδογόνο, μια εξωκυττάρια πρωτεΐνη, η οποία δεσμεύεται εκλεκτικά στον υποδοχέα GP IIb/IIIa. Το τριπεπτίδιο RGD (Arg-Gly-Asp) συνιστά τη μικρότερη αλληλουχία, η οποία είναι απαραίτητη για την αναγνώριση και πρόσδεση του ινωδογόνου στον υποδοχέα και απαντάται και σε άλλες συγκολλητικές πρωτεΐνες, οι οποίες είναι παρούσες στον εξωκυττάριο χώρο και στο αίμα, όπως η ινοσυνδετίνη, το κολλαγόνο, ο παράγοντας Von Willebrand, κτλ. Η αντιπηκτική θεραπεία έχει βασιστεί σε δύο διαφορετικές προσεγγίσεις του προβλήματος. Η μία προσέγγιση αφορά την εμπόδιση της πρωταρχικής διέγερσης των αιμοπεταλίων από διάφορους αγωνιστές, όπως θρομβίνη, επινεφρίνη, κολλαγόνο, κτλ. Η άλλη προσέγγιση περιλαμβάνει την διακοπή του μηχανισμού μεταγωγής σήματος, ο οποίος ακολουθεί την πρόσδεση του αγωνιστή στην επιφάνεια των αιμοπεταλίων. Η ασπιρίνη, παράγωγο του σαλικυλικού οξέος, αναστέλλει το πρώτο βήμα στη βιοσύνθεση της θρομβοξάνης Α2 από αραχιδονικό οξύ μέσω ακετυλίωσης του ενζύμου κυκλοοξυγενάση 1. Στην παρούσα διατριβή πραγματοποιήθηκε ο σχεδιασμός και η σύνθεση γραμμικών και κυκλικών αναλόγων του τριπεπτιδίου RGD με ενσωματωμένο σαλικυλικό οξύ ή παράγωγά του. Τα διάφορα ανάλογα συντέθηκαν με κλασικές μεθόδους πεπτιδικής σύνθεσης σε υγρή και στερεά φάση. Τη σύνθεση των αναλόγων ακολούθησε καθαρισμός τους (HPLC) και προσδιορισμός της δομής τους με (ESI-MS). Στη συνέχεια, προσδιορίστηκε in vitro με φωτομετρική μέθοδο στους 37C και συνεχή καταγραφή της διερχόμενης ακτινοβολίας με ειδικό όργανο (Dual Channel Aggregometer) η ανασταλτική τους δράση στη συγκολλητικότητα των αιμοπεταλίων του ανθρώπου. Προς περαιτέρω επιβεβαίωση των πειραμάτων συσσώρευσης και μελέτη της πρόσδεσης των αναλόγων στις ιντεγκρίνες χρησιμοποιήθηκε η κυτταρομετρία ροής με μονοκλωνικά αντισώματα έναντι των υποδοχέων Gp Ia, Gp IIb/IIIa, Gp IIIa και GMp 140. Αναλύοντας τα αποτελέσματα των βιολογικών μελετών, τόσο της αναστολής της συσσωμάτωσης των αιμοπεταλίων του ανθρώπου in vitro όσο και της κυτταρομετρίας ροής σε ενεργοποιημένα αιμοπετάλια για τα δραστικά πεπτίδια, οδηγούμαστε στα επόμενα συμπεράσματα: 1. Από τη σειρά των RGD γραμμικών αναλόγων που μελετήθηκαν, βρέθηκαν δραστικά μόνο στην περίπτωση που τα πεπτίδια έχουν στο C-τελικό τους άκρο αμίδιο. 2. Η σύζευξη του σαλικυλικού οξέος στο τριπεπτίδιο - αμίδιο RGD ενισχύει την αντισυγκολλητική του δράση έναντι των αιμοπεταλίων in vitro. Από αυτά τα ανάλογα 26 (IC50= 50μΜ), 27 (38μΜ) και 28 (53μΜ) (ενσωματωμένο σαλικυλικό οξύ στο τριπεπτίδιο) έχουν την ισχυρότερη δράση, ενώ μόνο το τριπεπτίδιο 23 έχει IC50= 540μΜ 3. Η προστασία του β-καρβοξυλίου του Asp με βενζυλομάδα αυξάνει τη δράση του πεπτιδίου σε σχέση με την ύπαρξη ελεύθερου β-καρβοξυλίου. Αυτό διαπιστώνεται από το γεγονός ότι όλα τα βιολογικώς δραστικά ανάλογα έχουν το β-καρβοξύλιο προστατευμένο με βενζυλομάδα και αυτό έρχεται σε συμφωνία με βιβλιογραφικά δεδομένα άλλων ερευνητών περί αναγκαιότητας ύπαρξης λιπόφιλης ομάδας στο C-τελικό άκρο του πεπτιδίου. 4. Αντίθετα, η ενσωμάτωση σαλικυλο-παραγώγων (βρώμο-, χλώρο-, νίτρο-, άμινο-, κτλ) στα ανάλογα δίνει πολύ μικρή αντισυγκολλητική δράση στα αιμοπετάλια του ανθρώπου in vitro σε σχέση με το σαλικυλικό οξύ. 5. Από τα συντεθέντα κυκλικά ανάλογα μόνο το ανάλογο 61, που φέρει δισουλφιδικό δεσμό μεταξύ της κυστεΐνης και του θειοσαλικυλικού οξέος, επέδειξε ισχυρή αντισυγκολλητική δράση έναντι των αιμοπεταλίων του ανθρώπου in vitro με τιμή IC50= 8μΜ, που είναι και η καλύτερη τιμή IC50 για όλα τα ανάλογα που συντέθηκαν (γραμμικά και κυκλικά). 6. Και στην περίπτωση των κυκλικών πεπτιδίων, τα ανάλογα με το προστατευμένο β-καρβοξύλιο εμφανίζουν ισχυρότερη ανασταλτική δράση έναντι εκείνων που φέρουν το β-καρβοξύλιο ελεύθερο. 7. Από όλα τα γραμμικά ανάλογα που περιέχουν παράγωγα του σαλικυλικού οξέος το ανάλογο 39 που περιέχει το 5-χλωρο σαλικυλικό οξύ εμφανίζει ισχυρή ανασταλτική δράση έναντι του υποδοχέα Gp Ib. 8. Τέλος, θα πρέπει να αναφερθεί ότι είναι η πρώτη φορά που συνθετικά πεπτιδικά ανάλογα του RGD εμφανίζουν ισχυρή πρόσδεση στον υποδοχέα Gp Ib, o οποίος ευθύνεται για την προσκόλληση των αιμοπεταλίων στο κυτταρικό τοίχωμα. / Integrins constitute a large family of heterodimeric cell-surface, transmembrane receptors, which play a major role in cell/cell and cell/matrix adhesive interactions. The Arg-Gly-Asp (RGD) sequence is known to be the integrin recognition site of many extracellular matrix proteins such as fibronectin, osteopontin, collagen, fibrinogen, von Willebrand factor, laminin, etc. On the other hand, it is well known that low doses of aspirin (acetyl salicylic acid) decrease platelet aggregation by causing an inhibitory effect on thromboxane A2 production by platelets. Several antiplatelet strategies have already been developed and are under preclinical or clinical investigation. In the present thesis, the synthesis of linear and cyclic RGD analogs incorporating salicylic acid derivatives is reported. The syntheses of the new analogs were carried out by using classic methods of peptide synthesis in liquid or solid phase. The synthesized compounds were purified by RP-HPLC and lyophilised to give fluffy solid, identified by ESI-MS spectra. These compounds were tested for inhibitory activity on human platelet aggregation in vitro, by adding common aggregation reagents to citrated platelet rich plasma (PRP). The aggregation was determined using a dual channel electronic aggregometer by recording the increase of light transmission. Their specificity for the Gp receptors was checked by using flow cytometry with monoclonal antibodies against Gp Ib, Gp IIb/IIIa, Gp IIIa and GMP140 receptors. Based on the results of the biological studies we could report the next inferences: 1. From the studied synthetic RGD analogs only peptides – amides are active against human platelet aggregation in vitro. 2. The coupling of salicylic acid with the RGD peptides enforces the antiplatelet activity in vitro of the single tripeptide. From the above peptides, the analog 26 (tripeptide incorporating salicylic acid) shows strong antiplatelet activity (IC50=50 μΜ), whereas the analog 23 (only tripeptide) has IC50= 540μΜ. 3. The protection of the β-carboxy group of Asp as benzylester increases the activity of the peptides in comparison with those having the β-carboxy group unprotected. Thus, our results ensure the theory of necessity of the existence a lipophile center on the C-terminal side of the peptide. 4. The incorporation of salicylic acid derivatives in the RGD peptide does not increase further the antiplatelet activity than the incorporation of salicylic acid does. 5. Among the cyclic RGD peptides only the analog 61, having the disulfide bridge between the cysteine and the thiosalicylic acid, shows strong antiplatelet activity in vitro (IC50= 8μΜ). 6. Most of the analogs show high binding affinity for the Gp Ib receptor. The cyclic analog 61 shows special selectivity for this receptor at concetrations of 110 μΜ. 7. The analog 39, although it shows low antiplatelet activity, has high binding affinity for the Gp Ib receptor. Probably, this activity is due to the atom of Cl at the 5 position of aromatic ring of salicylic acid. 8. According to the literature data, it is the first time that synthetic RGD peptides show strong binding affinity for the Gp Ib receptor, which is responsible for the platelet adhesion to the subenthothelium.

Les techniques de visualisation de traces sont fréquemment employées par les développeurs pour comprendre, déboguer, et optimiser leurs applications.La plupart des outils d'analyse font appel à des représentations spatiotemporelles, qui impliquent un axe du temps et une représentation des ressources, et lient la dynamique de l'application avec sa structure ou sa topologie.Toutefois, ces dernières ne répondent pas au problème de passage à l'échelle de manière satisfaisante. Face à un volume de trace de l'ordre du Gigaoctet et une quantité d'évènements supérieure au million, elles s'avèrent incapables de représenter une vue d'ensemble de la trace, à cause des limitations imposées par la taille de l'écran, des performances nécessaires pour une bonne interaction, mais aussi des limites cognitives et perceptives de l'analyste qui ne peut pas faire face à une représentation trop complexe. Cette vue d'ensemble est nécessaire puisqu'elle constitue un point d'entrée à l'analyse~; elle constitue la première étape du mantra de Shneiderman - Overview first, zoom and filter, then details-on-demand -, un principe aidant à concevoir une méthode d'analyse visuelle.Face à ce constat, nous élaborons dans cette thèse deux méthodes d'analyse, l'une temporelle, l'autre spatiotemporelle, fondées sur la visualisation. Elles intègrent chacune des étapes du mantra de Shneiderman - dont la vue d'ensemble -, tout en assurant le passage à l'échelle.Ces méthodes sont fondées sur une méthode d'agrégation qui s'attache à réduire la complexité de la représentation tout en préservant le maximum d'information. Pour ce faire, nous associons à ces deux concepts des mesures issues de la théorie de l'information. Les parties du système sont agrégées de manière à satisfaire un compromis entre ces deux mesures, dont le poids de chacune est ajusté par l'analyste afin de choisir un niveau de détail. L'effet de la résolution de ce compromis est la discrimination de l'hétérogénéité du comportement des entités composant le système à analyser. Cela nous permet de détecter des anomalies dans des traces d'applications multimédia embarquées, ou d'applications de calcul parallèle s'exécutant sur une grille.Nous avons implémenté ces techniques au sein d'un logiciel, Ocelotl, dont les choix de conception assurent le passage à l'échelle pour des traces de plusieurs milliards d'évènements. Nous proposons également une interaction efficace, notamment en synchronisant notre méthode de visualisation avec des représentations plus détaillées, afin de permettre une analyse descendante jusqu'à la source des anomalies. / Trace visualization techniques are commonly used by developers to understand, debug, and optimize their applications.Most of the analysis tools contain spatiotemporal representations, which is composed of a time line and the resources involved in the application execution. These techniques enable to link the dynamic of the application to its structure or its topology.However, they suffer from scalability issues and are incapable of providing overviews for the analysis of huge traces that have at least several Gigabytes and contain over a million of events. This is caused by screen size constraints, performance that is required for a efficient interaction, and analyst perceptive and cognitive limitations. Indeed, overviews are necessary to provide an entry point to the analysis, as recommended by Shneiderman's emph{mantra} - Overview first, zoom and filter, then details-on-demand -, a guideline that helps to design a visual analysis method.To face this situation, we elaborate in this thesis several scalable analysis methods based on visualization. They represent the application behavior both over the temporal and spatiotemporal dimensions, and integrate all the steps of Shneiderman's mantra, in particular by providing the analyst with a synthetic view of the trace.These methods are based on an aggregation method that reduces the representation complexity while keeping the maximum amount of information. Both measures are expressed using information theory measures. We determine which parts of the system to aggregate by satisfying a trade-off between these measures; their respective weights are adjusted by the user in order to choose a level of details. Solving this trade off enables to show the behavioral heterogeneity of the entities that compose the analyzed system. This helps to find anomalies in embedded multimedia applications and in parallel applications running on a computing grid.We have implemented these techniques into Ocelotl, an analysis tool developed during this thesis. We designed it to be capable to analyze traces containing up to several billions of events. Ocelotl also proposes effective interactions to fit with a top-down analysis strategy, like synchronizing our aggregated view with more detailed representations, in order to find the sources of the anomalies.

O presente trabalho tem por objetivo estudar o cotidiano dos agregados sociais da Praça da Alfândega localizada no centro da cidade de Porto Alegre no Rio Grande Sul. Foram investigados a partir deste espaço os aspectos relacionais, identitários, as expressões subjetivas e estéticas, as fronteiras de convivências assim como as territorializações empreendidas pelos indivíduos que compõem estes agregados sociais no espaço da Praça da Alfândega. Além disso, esta dissertação tem como intuito discutir as questões pertinentes sobre a função do espaço público, cidadania e dos aspectos sobre as formas de sociabilidade e de insociabilidade nas metrópoles contemporâneas. Pode se dizer que a Praça da Alfândega, nesse mosaico de representações sociais, territórios justapostos e signos compreendem um importante lócus de análise do espaço urbano, motivo pelo fato que ela se apresenta para as formas sociais, denominado de agregados sociais um espaço dicotômico da sobrevivência material e da convivência social. Por esse motivo, o estudo do cotidiano dos agregados sociais presentes no espaço da Praça da Alfândega constituem também uma investigação sobre os fenômenos urbanos contemporâneos e suas múltiplas e emergentes práticas sociais se apresentam nos estilos de vida, das formas simbólicas e nos usos dos espaços nas grandes metrópoles. / This paper aims to study the social aggregations’ every day life from Praça da Alfândega, located in the center of Porto Alegre city in Rio Grande Sul. The aspects of relationship and identity, the subjective and esthetic expressions, the borders of acquaintance as well as the territories settled by people who compounded these social groups in the space of Praça da Alfândega were investigated. Besides, this paper also aims to discuss relevant questions about the function of public space, citizenship and aspects about the forms of sociability and unsociability in modern metropolis. It is possible to affirm that Praça da Alfândega, on this mosaic of social representation adjacent territories and symbols form an important locus of analysis about urban space. For this reason, it becomes a dichotomic space both of material survival and of social acquaintance for the social forms, called social aggregations. Because of this reason, the study of the social aggregations’ every day life settled in Praça da Alfândega constitutes also an investigation about recent urban phenomena and their several and emergent social practices according to ways of life, symbolical forms, and settlement in big cities.

Cette thèse traite la problématique de la dissémination et l’agrégation des données dans un contexte de réseaux VANET (Vehicle Ad-Hoc Networks). Cette problématique est fort intéressante, toujours d’actualité dans un monde de plus en plus urbanisé. En effet, d’un côté la dissémination permet d’informer les véhicules mobiles des principaux événements en temps utile, et de l’autre côté l’agrégation permet de résumer plusieurs données émanant de sources différentes concernant le même événement. Le challenge de la dissémination consiste à calculer la zone de relevance d’un événement, de délivrer les messages aux véhicules de cette zone, et de continuer à délivrer les messages en continu aux véhicules de cette zone. Le challenge de l’agrégation consiste essentiellement à sélectionner les messages à agréger et à qualifier les messages provenant de véhicules lointains. Pour résoudre le problème de dissémination nous proposons un nouveau protocole de dissémination des données dans les réseaux VANET. La principale idée de ce protocole est basée sur la définition de zones de relevance ZOR (zone of relevance of a région) pour la mesure de l’intérêt d’une zone par rapport à un évènement donné, et la définition de split Map permettant de décomposer une grande région en un ensemble de ZORs. L’approche de calcul des ZORs est formalisée, elle est basée sur les techniques de greedy pour l’extraction de la couverture pertinente. Le protocole de dissémination présenté sous forme de diagramme Flowchart qui résumé les activités lorsque qu’un véhicule est en mouvement, un événement est détecté. La performance du protocole proposé est évaluée et comparé au protocole Slotted1-Persistence à travers un environnement de simulations et une topologie réelle de routes de la ville de Bizerte en Tunisie. Les résultats de simulation sont présentés et discutés.D’autre part, certaines applications VANET, par exemple le système d’information de trafic (TIS), nécessitent une agrégation de données pour informer les véhicules des conditions de circulation, ce qui réduit les embouteillages et par conséquent les émissions de CO2 Par conséquent, la conception d'un protocole d'agrégation efficace combinant des informations de trafic corrélées telles que l'emplacement, la vitesse et la direction, appelées données flottantes sur les voitures (FCD), pose un problème complexe. Dans cette thèse, nous introduisons un nouveau protocole d’agrégation de données dans un réseau VANET appelé SDDA (Smart Directional Data Aggregation). Ce protocole est dédié aussi bien à l’échange de données dans un contexte urbain et autoroutier. Le protocole proposé est basé sur une sélection des messages à agréger. Trois principaux filtres ont été utilisés : filtrage basé sur la direction des véhicules, filtrage basé sur la limitation de vitesse, et filtrage basé sur l’élimination des messages dupliqués. Trois algorithmes d’agrégation sont proposés, ils visent à optimiser l’algorithme de SOTIS. Les trois algorithmes traitent des cas de routes unidirectionnelles, bidirectionnelles et les réseaux urbains. A l’image du chapitre précédent, la performance des algorithmes proposés sont évaluées à travail un travail de simulation et différents résultats sont présentés et discutés. / Since the last decade, the emergence of affordable wireless devices in vehicle ad-hoc networks has been a key step towards improving road safety as well as transport efficiency. Informing vehicles about interesting safety and non-safety events is of key interest. Thus, the design of an efficient data dissemination protocol has been of paramount importance. A careful scrutiny of the pioneering vehicle-to-vehicle data dissemination approaches highlights that geocasting is the most feasible approach for VANET applications, more especially in safety applications, since safety events are of interest mainly to vehicles located within a specific area, commonly called ZOR or Zone Of Relevance, close to the event. Indeed, the most challenging issue in geocast protocols is the definition of the ZOR for a given event dissemination. In this thesis, our first contribution introduces a new geocast approach, called Data Dissemination Protocol based on Map Splitting(DPMS). The main thrust of DPMS consists of building the zones of relevance through the mining of correlations between vehicles’ trajectories and crossed regions. To do so, we rely on the Formal Concept Analysis (FCA), which is a method of extracting interesting clusters from relational data. The performed experiments show that DPMS outperforms its competitors in terms of effectiveness and efficiency. In another hand, some VANET applications, e.g., Traffic Information System (TIS), require data aggregation in order to inform vehicles about road traffic conditions, which leads to reduce traffic jams and consequently CO2 emission while increasing the user comfort. Therefore, the design of an efficient aggregation protocol that combines correlated traffic information like location, speed and direction known as Floating Car Data (FCD) is a challenging issue. In this thesis, we introduce a new TIS data aggregation protocol called Smart Directional Data Aggregation (SDDA) able to decrease the network overload while obtaining high accurate information on traffic conditions for large road sections. To this end, we introduce three levels of messages filtering: (i) filtering all FCD messages before the aggregation process based on vehicle directions and road speed limitations, (ii) integrating a suppression technique in the phase of information gathering in order to eliminate the duplicate data, and (iii) aggregating the filtered FCD data and then disseminating it to other vehicles. The performed experiments show that the SDDA outperforms existing approaches in terms of effectiveness and efficiency.

This dissertation comprises of three original contributions to empirical forecasting research. Chapter 1 introduces the dissertation. Chapter 2 contributes to the literature on hierarchical time series (HTS) modelling by proposing a disaggregated forecasting system for both inflation rate and its volatility. Using monthly data that underlies the Retail Prices Index for the UK, we analyse the dynamics of the inflation process. We examine patterns in the time-varying covariation among product-level inflation rates that aggregate up to industry-level inflation rates that in turn aggregate up to the overall inflation rate. The aggregate inflation volatility closely tracks the time path of this covariation, which is seen to be driven primarily by the variances of common shocks shared by all products, and by the covariances between idiosyncratic product-level shocks. We formulate a forecasting system that comprises of models for mean inflation rate and its variance, and exploit the index structure of the aggregate inflation rate using the HTS framework. Using a dynamic model selection approach to forecasting, we obtain forecasts that are between 9 and 155 % more accurate than a SARIMA-GARCH(1,1) for the aggregate inflation volatility. Chapter 3 is on improving forecasts using forecast combinations. The paper documents the software implementation of the open source R package for forecast combination that we coded and published on the official R package depository, CRAN. The GeomComb package is the only R package that covers a wide range of different popular forecast combination methods. We implement techniques from 3 broad categories: (a) simple non-parametric methods, (b) regression-based methods, and (c) geometric (eigenvector) methods, allowing for static or dynamic estimation of each approach. Using S3 classes/methods in R, the package provides a user-friendly environment for applied forecasting, implementing solutions for typical issues related to forecast combination (multicollinearity, missing values, etc.), criterion-based optimisation for several parametric methods, and post-fit functions to rationalise and visualise estimation results. The package has been listed in the official R Task Views for Time Series Analysis and for Official Statistics. The brief empirical application in the paper illustrates the package’s functionality by estimating forecast combination techniques for monthly UK electricity supply. Chapter 4 introduces HTS forecasting and forecast combination to a healthcare staffing context. A slowdown of healthcare budget growth in the UK that does not keep pace with growth of demand for hospital services made efficient cost planning increasingly crucial for hospitals, in particular for staff which accounts for more than half of hospitals’ expenses. This is facilitated by accurate forecasts of patient census and churn. Using a dataset of more than 3 million observations from a large UK hospital, we show how HTS forecasting can improve forecast accuracy by using information at different levels of the hospital hierarchy (aggregate, emergency/electives, divisions, specialties), compared to the naïve benchmark: the seasonal random walk model applied to the aggregate. We show that forecast combination can improve accuracy even more in some cases, and leads to lower forecast error variance (decreasing forecasting risk). We propose a comprehensive parametric approach to use forecasts in a nurse staffing model that has the aim of minimising cost while satisfying that the care requirements (e.g. nurse hours per patient day thresholds) are met.

Background: Selective serotonin reuptake inhibitor (SSRI) antidepressants prevent serotonin (5-HT) uptake by the serotonin transporter (SERT). Since blood platelets express SERT, SSRIs may modify platelet function and the risk of cardiovascular disease. However, the beneficial or adverse effects of SSRIs on arterial thrombosis are poorly characterised and detailed in vitro experimental data is limited. The SSRI citalopram is a racemate, the (S)-isomer being the more potent SERT inhibitor. Although citalopram has been shown to inhibit platelets in vitro, it is unclear whether this is mediated via SERT blockade. Aim: To determine if citalopram inhibits platelet function via SERT blockade, or through a novel mechanism of action. Findings: 5-HT uptake into platelets was blocked by both citalopram isomers at concentrations that had no apparent effect on platelet function. Despite the (S)-citalopram isomer being the more potent SERT inhibitor, (R)-citalopram was equally potent at inhibiting other platelet functions. These findings strongly suggest that inhibition of platelet function by citalopram in vitro is not mediated by blocking SERT. Subsequent experiments identified two putative mechanisms for citalopram-mediated platelet inhibition: 1) citalopram did not inhibit calcium store release induced by the platelet agonist U46619, despite blocking subsequent Rap1 activation. A credible target for this inhibitory mechanism is the calcium and diacylglycerol guanine nucleotide exchange factor-1 (CalDAG-GEFI): 2) citalopram suppressed early protein phosphorylation within the GPVI pathway, resulting in the inhibition of subsequent platelet responses. Further experiments show that other commonly used antidepressants also inhibit platelets. As with citalopram, inhibition was only observed at concentrations above those required to block SERT, suggesting that alternative inhibitory mechanism(s) are responsible. Conclusions: Data presented in this thesis support two novel putative mechanisms of citalopram-induced platelet inhibition. These findings demonstrate that citalopram and other antidepressants inhibit platelets independently of their ability to block SERT-dependent 5-HT transport. The identification of thesemechanisms provides a pharmacological approach to develop novel antiplatelet agents based on current antidepressants.

The marine fishery in Kenya is predominantly small-scale and artisanal with about 11,000 fishers intensely fishing near shore coastal reefs using minimally selective fishing gears. A large majority (88%) of fishers use outdated equipment such as basket traps, beach seines, hand lines (hook and lines), fence traps, gillnets, spearguns and cast nets. Handmade canoes propelled by paddles (kasia) or sail power are used to access offshore waters, while only a few fishers have motorized boats. Although fishers along this coast know and express the potential of offshore fishing, most of them are disempowered and unable to access any of the largely untapped offshore pelagic resources. Using a unique dataset from four distinct coastal areas: Funzi-Shirazi bay area, Diani-Chale area, Mombasa-Kilifi north coast area and the Malindi-Ungwana bay area, containing species level length frequency catch data from the multi-gear and multi-species fishery, abundance of specific species, gear use comparisons in various regions, catch per unit effort and total catch estimate over a nine year period (2001 – 2009) were evaluated. Despite high diversity in the fishery, five species (Lethrinus lentjan, Siganus sutor, Leptoscarus vaigiensis, Lethrinus harak and Parupeneus macronemus) represented over 75% of the catch. A total of 11 legitimate gears were observed in this coastal artisanal fishery with basket traps (42%) being the most popular. Fishers along the Mombasa-Kilifi area predominatly used beach seines while those in Diani-Chale, Malindi-Ungwana bay and Funzi-shirazi bay predominaltly used spearguns, gillnets and basket traps, respectively. Apart from gillnets, a general declining trend for most of the gear types was observed since 2004. Beach seines recorded the lowest (20.9±0.2 cm) mean length while gillnets recorded the highest (34.2±0.3 cm). The highest catch (~26,000 metric tons) came in 2001 and the lowest (~15,000 metric tons) in 2005. The highest number of fishers was observed in 2008 while 2009 recorded the highest (4.8±2.3) mean number of hours per outing. The mean annual CPUE per region ranged from (1.5 kg.fisher-1.hr-1) in Diani-Chale to (1.0 kg.fisher-1.hr-1) in Malindi-Ungwana bay. Making use of questionnaire data, the attitudes towards offshore fishing strategies, FADs in particular, were evaluated. Some communities (about 25% in every location) were not even aware of FAD fisheries. With the imminent introduction of a FAD fishery in Kenya, it was concluded that, for this fishery to realize its full potential, training on FAD fishing techniques has to be done. Finally, effective management is necessary if small-scale fisheries are to continue providing food security for many poor coastal communities. Gear-based management in Kenya, although under represented and under studied, has the potential to be adaptive, address multiple objectives, and be crafted to the socio-economic setting. Management effectiveness in near shore fisheries has generally been evaluated at the scale of the fish community. However, community level indicators can mask species-specific declines that provide significant portions of the fisheries yields and income. This thesis seeks to identify ways in which the Kenyan artisanal fishery can be sustained and managed from within coastal communities, giving them the resources and education to effectively improve their lives. The introduction of a offshore FAD fishery and hence access to offshore pelagic species provides an opportunity to not only alleviate pressure on coastal resources but also to empower coastal communities and contribute to the growth of Kenya’s national economy as a whole.