Dissertations / Theses on the topic 'Profiling'

Memory profiling is an important technique which aids program optimization and can even help tracking down bugs. The main problem with the current memory profiling techniques and tools is that they slow down the target software considerably therefore making them inadequate for mainline integration. Ideally, the user would be able to monitor memory consumption without having to worry about the rest of the software being affected in any way. This thesis provides a comparison of existing techniques and tools along with the description of a memory profiler implementation which tries to provide a balance between the information it is able to retrieve and the influence it has on the target software.

Nonostante l'elevato numero di dati pubblicati come LD, il loro utilizzo non ha ancora mostrato il loro potenziale per l’assenza di comprensione dei metadati. I consumatori di dati hanno bisogno di ottenere informazioni dai dataset in modo veloce e concentrato per poter decidere se sono utili per il loro problema oppure no. Le tecniche di profilazione dei dati offrono una soluzione efficace a questo problema in quanto sono utilizzati per generare metadati e statistiche che descrivono il contenuto dei dataset. Questa tesi presenta una ricerca, che affronta i problemi legati alla profilazione Linked Data. Nonostante il termine profilazione dei dati è usato in modo generico per diverse informazioni che descrivono i dataset, in questa tesi noi andiamo a ricoprire tre aspetti della profilazione; topic-based, schema-based e linkage-based. Il profilo proposto in questa tesi è fondamentale per il processo decisionale ed è la base dei requisiti che portano verso la comprensione dei dataset. In questa tesi presentiamo un approccio per classificare automaticamente insiemi di dati in una delle categorie utilizzate nel mondo dei LD. Inoltre, indaghiamo il problema della profilazione multi-topic. Per la profilazione schema-based proponiamo un approccio riassuntivo schema-based, che fornisce una panoramica sui rapporti nei dati. I nostri riassunti sono concisi e chiari sufficientemente per riassumere l'intero dataset. Inoltre, essi rivelano problemi di qualità e possono aiutare gli utenti nei compiti di formulazione dei query. Molti dataset nel LD cloud contengono informazioni simili per la stessa entità. Al fine di sfruttare appieno il suo potenziale LD bisogna far vedere questa informazione in modo esplicito. Profiling Linkage fornisce informazioni sul numero di entità equivalenti tra i dataset e rivela possibili errori.Le tecniche di profiling sviluppate durante questo lavoro sono automatiche e possono essere applicate a differenti insiemi di dati indipendentemente dal dominio.
Recently, the increasing diffusion of Linked Data (LD) as a standard way to publish and structure data on the Web has received a growing attention from researchers and data publishers. LD adoption is reflected in different domains such as government, media, life science, etc., building a powerful Web available to anyone. Despite the high number of datasets published as LD, their usage is still not exploited as they lack comprehensive metadata. Data consumers need to obtain information about datasets content in a fast and summarized form to decide if they are useful for their use case at hand or not. Data profiling techniques offer an efficient solution to this problem as they are used to generate metadata and statistics that describe the content of the dataset. Existing profiling techniques do no cover a wide range of use cases. Many challenges due to the heterogeneity nature of Linked Data are still to overcome. This thesis presents the doctoral research which tackles the problems related to Profiling Linked Data. Even though the term of data profiling is the umbrella term for diverse descriptive information that describes a dataset, in this thesis we cover three aspects of profiling; topic-based, schema-based and linkage-based. The profile provided in this thesis is fundamental for the decision-making process and is the basic requirement towards the dataset understanding. In this thesis we present an approach to automatically classify datasets in one of the topical categories used in the LD cloud. Moreover, we investigate the problem of multi-topic profiling. For the schema-based profiling we propose a schema-based summarization approach, that provides an overview about the relations in the data. Our summaries are concise and informative enough to summarize the whole dataset. Moreover, they reveal quality issues and can help users in the query formulation tasks. Many datasets in the LD cloud contain similar information for the same entity. In order to fully exploit its potential LD should made this information explicit. Linkage profiling provides information about the number of equivalent entities between datasets and reveal possible errors. The techniques of profiling developed during this work are automatic and can be applied to different datasets independently of the domain.

In this thesis, we discuss several profilers and use selected ones to optimize a seismic application for StatoilHydro, Norway's main oil company. Paralellization techniques are also discussed. The application scans multiple traces of seismic data and removes unwanted multiples(noise). Seismic applications are central in simulations aiding geophysicists in finding oil reservoirs. The motivation for selecting this particular application, Adafil, is that it needs to be faster to be useful in practice. Abstract Our work gives several useful general hints for how to parallelize and optimize such applications for modern architectures. Abstract The application is profiled using several tools, singeling out three hotspots. This thesis will show that this application has some L2 cache misses. which can be avoided with prefetching. The work also shows that specific parts of the code, among others one containing a convolution algorithm, can benefit greatly by using FFT to lower complexity from O(n^2) to O(n log n) for these parts, and by leveraging the adaptive implementations of FFTW leads to a significant speedup of the application.

It was observed, that the RT-PICLS code ran by FWKT on the hypnos cluster was producing an unusual amount of system load, according to Ganglia metrics. Since this may point to an IO-problem in the code, this code was analyzed more closely.

We have introduced the notion of user profiling with trust, as a solution to theproblem of uncertainty and unmanageable exposure of personal data duringaccess, retrieval and consumption by web applications. Our solution sug-gests explicit modeling of trust and embedding trust metrics and mechanismswithin very fabric of user profiles. This has in turn allowed information sys-tems to consume and understand this extra knowledge in order to improveinteraction and collaboration among individuals and system. When formaliz-ing such profiles, another challenge is to realize increasingly important notionof privacy preferences of users. Thus, the profiles are designed in a way toincorporate preferences of users allowing target systems to understand pri-vacy concerns of users during their interaction. A majority of contributionsof this work had impact on profiling and recommendation in digital librariescontext, and was implemented in the framework of EU FP7 Smartmuseumproject. Highlighted results start from modeling of adaptive user profilesincorporating users taste, trust and privacy preferences. This in turn led toproposal of several ontologies for user and content characteristics modeling forimproving indexing and retrieval of user content and profiles across the plat-form. Sparsity and uncertainty of profiles were studied through frameworksof data mining and machine learning of profile data taken from on-line so-cial networks. Results of mining and population of data from social networksalong with profile data increased the accuracy of intelligent suggestions madeby system to improving navigation of users in on-line and off-line museum in-terfaces. We also introduced several trust-based recommendation techniquesand frameworks capable of mining implicit and explicit trust across ratingsnetworks taken from social and opinion web. Resulting recommendation al-gorithms have shown to increase accuracy of profiles, through incorporationof knowledge of items and users and diffusing them along the trust networks.At the same time focusing on automated distributed management of profiles,we showed that coverage of system can be increased effectively, surpassingcomparable state of art techniques. We have clearly shown that trust clearlyelevates accuracy of suggestions predicted by system. To assure overall pri-vacy of such value-laden systems, privacy was given a direct focus when archi-tectures and metrics were proposed and shown that a joint optimal setting foraccuracy and perturbation techniques can maintain accurate output. Finally,focusing on hybrid models of web data and recommendations motivated usto study impact of trust in the context of topic-driven recommendation insocial and opinion media, which in turn helped us to show that leveragingcontent-driven and tie-strength networks can improve systems accuracy forseveral important web computing tasks.

QC 20130219

In Chapter one I introduce the subject of geographic profiling, its use in criminology and its previous application to biology. I go on in Chapter two to examine the original model and develop a likelihood-based approach to fit the parameters to data from 53 UK invasive species. GP performs well on this novel problem, and outperforms other simple spatial modelling techniques. Using simulations I show that GP is particularly efficient at locating sources when there is more than a single source. Chapter three develops a Bayesian approach using Dirichlet Processes to account for the problem of multiple sources. This model was developed in collaboration with Robert Verity. This new Bayesian model outperforms the original model used in criminology and offers a range of additional information from the data. The Bayesian GP model is then used to determine the sources of malaria outbreaks in Cairo. These developments significantly improve and extend the theory and application of GP. In Chapter four I discuss the possible shapes of dispersal functions. I conduct a review of the literature and find a geometric mistake in the way linear distributions have been extracted from two-dimensional data. The correct back-transformation allows these dispersal distributions to be properly generated. Using this information; ecologists, conservationists and resources managers can now apply GP to real world problems and effectively allocate limited resources to locate sources of species invasions and disease outbreaks. I go on in Chapter five to develop a method for fitting the primary parameter sigma from the point pattern data and run simulations to show the effectiveness of this new approach. In Chapter six I illustrate the application of GP to three problems, one in criminology, one in ecology and one in epidemiology. I finish by summarising the work in this thesis and discussing the potential future developments and applications of GP.

Throughout this thesis, metabolic profiling using a global mass spectrometry-based approach was utilised to explore cholangiocarcinoma (CCA) metabolic signature in urine and serum. CCA is an aggressive disease with a poor prognosis. The geographical distribution of CCA cases is markedly uneven and it closely mirrors the prevalence of its predisposing risk factors. CCA is most commonly seen in Southeast Asia where liver fluke infection is endemic, whereas in Western countries it is a rare malignancy with an unknown aetiology in most cases. To date, a multi-disciplinary approach with multi-modal tools are often applied for the diagnosis of this complex disease. The primary aim of this thesis was to provide insights into the perturbed biological pathways underlying cholangiocarcinogenesis as a mean to identify metabolite biomarkers to improve CCA screening and diagnosis. The first four results chapters describes the urinary metabolome associated with liver fluke induced-CCA and sporadic CCA in patients from Thailand and the UK, respectively. Overall, CCA induces a metabolic profile change that can distinguish patients from healthy and control groups with various hepatobiliary pathologies. Dysregulation in acylcarnitine, bile acid, steroid and purine metabolism was characteristic of the CCA urinary metabolic signature in both populations. A panel of 10 metabolites (primarily acylcarnitine and steroid species) achieved a diagnostic accuracy of 93.4% and AUC value of 98.8% (CI= 96.3%-100%) in distinguishing CCA cases from healthy participants. Phenotypic disparities between the two distinct populations were primarily related to differences in dietary practices and body composition. Furthermore, regardless of the underlying aetiology, biological perturbations associated with the CCA urine metabolome signature appeared to be influenced by gut microbial community metabolism. Sporadic bile duct disease exhibited altered nucleotide metabolism, possibly indicating differences in mitochondrial energy production pathways related to differences in CCA underlying aetiology. Similarly, the serum metabolic patterns were differential in bile duct cancer, compared to healthy participants and individuals with benign and malignant hepatobiliary diseases. Impaired phospholipid homoeostases was particularly observed in CCA serum profiles, possibly suggesting influence of genetic variants that impair biliary phospholipid secretion on circulating lipids. A common metabotype, primarily indicative of inflammatory signature, was found to be associated with liver pathologies, both benign and malignant. Overall, it is possible to conclude that metabonomic profiling can identify disease-related metabotypes and may possibly contribute to uncovering novel candidate diagnostic biomarkers in CCA.

Thesis (M.B.A.)--Massachusetts Institute of Technology, Sloan School of Management, 2004.
"June 2004."
Includes bibliographical references (leaves 34-35).
Analysis of new and renewing microborrower risk behaviors at the Mexican microfinance institution FinComun shows that risk behavior can be predicted. The range of risk behaviors analyzed include: incidence of late payments, maximum number of weeks in arrears, and default. Males generally exhibited somewhat riskier patterns of credit behavior than females. Tools are provided for defining and predicting risky credit behavior. The implication is that these tools can be used to improve the management of credit risk and, therefore, improve the effectiveness of the microfinance program.
by Colin D. Nicholson.
M.B.A.

L’évolution de l’Internet dans les dernières années a été caractérisée par des changement dramatiques dans la manière dont les utilisateurs se comportent, interagissent et utilisent le réseau. Ceci a été particulièrement accompagné par l’introduction de nouvelles classes d’applications telles que les jeux en ligne et les réseaux pair-à-pair. L’un des défis les plus importants pour les administrateurs réseau et les ISPs est alors devenu l’identification du trafic Internet afin de pouvoir protéger leurs ressources contre le trafic indésirable et de prioriser certaines applications majeures. Les méthodes statistiques sont préférées à celles basées sur le numéro de port et l’inspection approfondie des paquets, car elles sont robustes au changement malveillant du numéro de port et fonctionnent avec le trafic crypté. Ces méthodes combinent l’analyse des paramètres statistiques des flux de paquets, tels que la taille des paquets et le temps les séparant, avec des techniques issues de la théorie d’apprentissage (machine learning). La majorité des méthodes statistiques ne peuvent pas identifier les flux applicatifs en temps réel et elles ont besoin d’atteindre la fin des flux avant de prendre une décision sur leur nature. Ceci est considéré comme trop long pour la plupart des administrateurs réseau, puisqu’il ne permet pas de bloquer un flux Internet indésirable à son début ni de lui donner en amont une qualité particulière de service. Un autre défi important pour les administrateurs réseau est de détecter et diagnostiquer tout changement dans le réseau comme une congestion à long terme, un changement dans le routage, une défaillance d’une liaison ou tout autre événement entraînant un changement dans les délais réseau. Dans la littérature, il y a un grand nombre de méthodes pour détecter des anomalies dans le réseau, mais la plupart de ces méthodes ont besoin de générer un volume considérable de trafic destiné) la métrologie du réseau. La réduction de la charge des mesures est un besoin vital pour les administrateurs réseaux. Dans cette thèse, nous décrivons les travaux que nous avons menés sur l’identification du trafic Internet et sur la détection des anomalies dans les réseaux. Dans la première partie, nous présentons nos trois méthodes que nous avons développées au cours de cette thèse, et qui permettent d’identifier avec précision et à la volée le trafic Internet. La première méthode, par sa nature itérative et probabiliste, identifie les applications rapidement et avec une grande précision en utilisant uniquement la taille des N premiers paquets. La deuxième méthode enrichit la première avec les temps entre paquets, pour cela nous avons eu besoin d’introduire un modèle pour filtrer le bruit dû aux conditions du réseau et d’extraire des mesures le temps d’attente due aux applications. Notre troisième méthode pour la classification du trafic en ligne combine les approches statistiques à des informations sur le comportement des machines hôtes afin de rendre l’identification du trafic Internet encore plus précis tout en profilant les activités réseaux des hôtes. Pour notre troisième méthode, nous utilisons la taille des paquets comme paramètre principal et nous exploitons les informations sur l’interaction des machines pour mieux affecter un flux à une application. Dans la deuxième partie de cette thèse, nous abordons le problème de détection des anomalies dans les réseaux. Nous commençons par une étude sur la stabilité des systèmes de coordonnées Internet (particulier Vivaldi). Dans une première étape, nous confirmons le fait que les coordonnées de Vivaldi oscillent au fil du temps en raison de la nature adaptative du système. Toutefois, les variations de ces coordonnées sont dans la plupart des temps en corrélation les unes avec les autres, pointant par conséquent vers un cluster de nœuds stables vu de l’intérieur du réseau. Dans un deuxième temps, nous présentons un nouvel algorithme de cloustering basé sur des méthodes de groupement hiérarchique afin d’identifier ce cluster de nœuds stables. Enfin, nous soulignons l’utilité d’une telle constatation avec une application qui permet de détecter les changements dans le réseau. En changeant artificiellement les délais du réseau dans différents scénarios, nous montrons que ces changements sont reflétés par ce corps de nœuds stables, permettant ainsi d’obtenir une image globale de la stabilité du réseau sans avoir besoin de mesures exhaustives des délais
The evolution of the Internet in the last few year has been characterized by dramatic changes in the way users behave, interact and utilize the network. This was accompanied by the introduction of new categories of applications such as network games and peer-to-peer services. One of the most important challenges for network administrators and ISPs is then becoming the identification of Internet traffic applications in order to protect their resources from unwanted traffic and to prioritize some major applications. Statistical methods are preferred to port-based ones and deep packet inspection since they donât rely on the port number and they also work for encrypted traffic. Theses methods combine the statistical analysis of the application packet flow parameters, such as packet size and inter-packet time, with machine learning techniques. However, the majority of these statistical methods cannot identify flows early and require reaching the end of flows before taking any decision which is considered as too late for network administrators ; indeed they do not provide means to stop an Internet flow or to give it a special quality of service early in its lifetime. Another important challenge for network administrators is to detect and diagnose key network changes as a long-term congestion, a rerouting, a link failure or any other even causing a shift in network delays. In the literature there is a huge amount of anomaly detection methods but most of them require exhaustive measurements to function properly. Reducing the load of network-wide monitoring is always a vital need for network administrators. In this thesis we present several contributions around Internet traffic identification and network-wide anomaly detection. In the first part we present three methods we have developed in order to identify accurately and on the fly the Internet traffic. The first method is a new online iterative probabilistic method that identifies applications quickly and accurately by only using the size of the first N packets. The second method enhances the first one with the inter-packet time in order to identify Internet traffic, this has required the introduction of a model to isolate the noise due to network conditions and to extract the time generated by the applications themselves. Our third method is a new online method for traffic classification that combines the statistical and host-based approaches in order to construct a robust and precise method for early Internet traffic identification. We use the packet size as the main feature for the classification and we benefit from the traffic profile of the host (i. E. Which application and how much) to decide in favour of this or that application. In the second part of this thesis, we aboard the problem of network-wide anomaly detection. We start by making a study about the stability of Internet coordinate systems (especially Vivaldi). In a first stage we confirm the fact that Vivaldi coordinates are most of the time correlated with each other pointing to a stable cluster of nodes seen from inside the network. In a second stage, we present a new clustering algorithm based on the data mining Hierarchical Grouping method to identify this cluster of stable nodes. Finally, we highlight the utility of such finding with and application that tracks changes in network delays. By changing artificially the network delays in different scenarios, we show that these changes are easily reflected by this body of stable nodes, hence allowing to obtain a global picture about the stability of the underlying network without the need for exhaustive delay measurements

Profiling of an application identifies parts of the code being executed using the hardware performance counters thus providing the application’s performance. Profiling has long been standard in the development process focused on a single execution of a single program. As computing systems have evolved, understanding the bigger picture across multiple machines has become increasingly important. As supercomputing grows in pervasiveness and scale, understanding parallel applications performance and utilization characteristics is critically important, because even minor performance improvements translate into large cost savings. The study surveys various tools for the application. After which, Perfminer was integrated in SCANIA’s Linux clusters to profile CFD and FEA applications exploiting the batch queue system features for continuous system wide profiling, which provides performance insights for high performance applications, with negligible overhead. Perfminer provides stable, accurate profiles and a cluster-scale tool for performance analysis. Perfminer effectively highlights the micro-architectural bottlenecks.
Profilering av en ansökan identifierar delar av koden exekveras med hjälp av hårdvara prestandaräknare därmed ger programmets prestanda. Profilering har länge varit standard i utvecklingsprocessen fokuserad på en enda exekvering av ett enda program. Som datorsystem har utvecklats, att förstå helheten på flera datorer har blivit allt viktigare. Som superdatorer växer i genomslagskraft och skala, är förståelsen parallella applikationer prestanda och användningsegenskaper avgörande betydelse, eftersom även prestandaförbättringar mindre översätta till stora kostnadsbesparingar. Studien granskar olika verktyg för tillämpningen. Därefter var Perfminer integrerat i Scanias Linux-kluster att profilera CFD och FEA-program som utnyttjar sats kösystem funktioner för kontinuerlig hela systemet profilering, vilket ger prestanda insikter för högpresterande tillämpningar, med försumbar overhead. Perfminer ger stabila, noggranna profiler och ett kluster skala verktyg för prestandaanalys. Perfminer belyser effektivt mikro arkitektoniska flaskhalsar.

Capturing the quality of software and detecting sections for further scrutiny within are of high interest for industry as well as for education. Project managers request quality reports in order to evaluate the current status and to initiate appropriate improvement actions and teachers feel the need of detecting students which need extra attention and help in certain programming aspects. By means of software measurement software characteristics can be quantified and the produced measures analyzed to gain an understanding about the underlying software quality.

In this study, the technique of code profiling (being the activity of creating a summary of distinctive characteristics of software code) was inspected, formulized and conducted by means of a sample group of 19 industry and 37 student programs. When software projects are analyzed by means of software measurements, a considerable amount of data is produced. The task is to organize the data and draw meaningful information from the measures produced, quickly and without high expenses.

The results of this study indicated that code profiling can be a useful technique for quick program comparisons and continuous quality observations with several application scenarios in both industry and education.

This thesis is about protein profiling in serum and plasma using antibody suspension bead arrays for the analysis of biobanked samples and in the context of prostate cancer biomarker discovery. The influence of sample preparation methods on antibody based protein profiles were investigated (Papers I-III) and a prostate cancer candidate biomarker identified and verified (Papers III-V). Furthermore, a perspective on the research area affinity proteomics and its’ employment in biomarker discovery, for improved understanding and potentially improved disease diagnosis, is provided. Paper I presents the results of a comparative plasma and serum protein profiling study, with a targeted biomarker discovery approach in the context of metabolic syndrome. The study yielded a higher number of significant findings and a low experimental variability in blood samples prepared as plasma. Paper II investigated the effects from post-centrifugation delays at different temperatures prior sample storage of serum and plasma samples. Minor effects were found on the detected levels of more than 300 predicted or known plasma proteins. In Paper III, the detectability of proteins in plasma was explored by exposing samples to different pre-analytical heat treatments, prior target capture. Heat induced epitope retrieval was observed for approximately half of the targeted proteins, and resulted in the discovery of different candidate markers for prostate cancer. Several antibodies towards the prostate cancer candidate biomarker CNDP1 were generated, epitope mapped and evaluated in a bead based sandwich immunoassay, as presented in Papers IV and V. Furthermore, the developed sandwich immunoassay targeting multiple distinct CNDP1 epitopes in more than 1000 samples, confirmed the association of CNDP1 levels to aggres- sive prostate cancer and more specifically to prostate cancer patients with regional lymph node metastasis (Paper V). As an outcome of the present investigations and in parallel to studies within the Biobank profiling research group, valuable lessons from study design and multiplex antibody analysis of plasma within biomarker discovery to experimental, technical and biological verifications have been collected.

QC 20130821

Effective dynamic virtual-machine optimization depends on quickly finding and optimizing frequently-executed code regions. Such regions are naturally represented with a trace-cache, a collection of long, frequently-executed sequences of instructions. An ideal trace-cache is small and captures the majority of the program's execution stream, yet consists of traces large enough to allow for useful inter-bytecode optimizations and improve instruction-fetch performance. This thesis describes an efficient dynamic technique for finding and constructing such traces.
Our trace caching algorithm relies on three innovative techniques: profiling data is managed with the Branch Implication Graph(BI-Graph) data structure, decomposing the program into small, independent, tree-like components using our hot subgraph filters the BI-Graph for relevant data, and traces are constructed from components using a minimum-bounded expected completion rate.
We evaluated the runtime performance of our generated traces using a trace cache simulator implemented on top of the Java Virtual Machine, SableVM. Tested against Java benchmarks, our primary data structure, the BI-Graph, scales linearly with the loaded program size and is no larger than a control flow graph. On our benchmarks set, our traces capture 98.5% of the executed instructions, average 34 bytecodes instructions in length, and execute to completion 98.5% of the time. Further, the total size of all traces created scales linearly at 83% of the program's size, averaging only 50000 bytecodes. The entire algorithm, including context-sensitive, branch bias profiling and trace creation, has a projected overhead of only 4.5%. Thus, our algorithm efficiently constructs a small trace cache consisting of long traces that both tend to execute to completion and capture the majority of the execution stream.

Bleomycin is a potent anti-tumour compound used in the treatment of squamous cell carcinomas. An unfortunate side effect of this drug is pulmonary toxicity. The onset of this damage manifests as mild oedema and inflammation which eventually develops into pulmonary fibrosis. The ability to correctly identify patients showing early signs of lung injury could significantly reduce the morbidity associated with bleomycin treatment. As such, this study was undertaken to identify genetic markers of early oedema and inflammation. A model of mild pulmonary injury was induced by bleomycin. Conventional quantitative analysis of broncho-alveolar lavage was used to indicate the severity of the oedematous response, whilst morphological changes were identified by histology and electron microscopy. Macroarrays were used to measure the expression of multiple genes during mild, progressive and severe oedema. Following normalisation and statistical analysis, gene expression patterns were compared from saline- and bleomycin-treated rats. A variety of genes were differentially expressed during each model, with the number increasing with the severity of the oedema. A cluster and two individual genes were consistently expressed across two of the models of oedema. The magnitude of the changes in gene expression were quantified and confirmed by quantitative PCR. In summary, complete toxicological and histological characterisation of the bleomycin-induced model of pulmonary injury successfully identified specific endpoints of injury. This model proved to be ideal for studying differential gene expression in response to drug-induced pulmonary oedema. A cluster of ion channels and trafficking genes has the potential to act as a biomarker. Two specific genetic markers (Na+/CI- betaine/GABA transporter, glucocorticoid receptor), and a protein marker (cocoacrisp) have been identified for the oedema. In addition to these genes and protein being potential biomarkers of injury, they are also prospective targets for clinical treatment.

In the last 20 years mobile devices gained an important role in daily life and became must have items for everyone. As mobile devices give us the much needed flexibility and mobility, they also represent one major concern; security. As the information is transmitted from node to node via radio frequencies, an imposter can gain access into a mobile network without the need to gain physical access to firewalls and gateways. Also, as they are light and small, mobile devices are easily lost and often used without any PIN or password protection enabled. Hence, it is not difficult for someone even without any technical knowledge to gain access to such devices if they have been left behind or stolen. As traditional intrusion detection systems are not very effective against this kind of attack, there is a need of a different approach that can assist in the identification of a potential imposter. This thesis begins by assessing the security needs of the mobile devices, and establishes the perceived inadequacy of existing safeguards in this respect. Therefore this research considers using Behaviour-Based Mobile Intrusion Detection System (BeMIDS) that aims to assist the identification of anomalous user activity. This in return presents the two main characteristics needed to classify a legitimate user inside a mobile network: first with whom, when and what type of connection is established and then at where the mobile device is left open. After this the research proposes a novel approach that investigates the application of three machine learning algorithms to profile user behaviour in mobile networks. In BeMIDS, historical user profiles are created and then compared with the real-time ones in order to detect unusual activity in mobile networks. If a user’s behaviour changes, this results in alerting the system as an anomalous activity. Specific examples of behaviours that BeMIDS appears to be particularly sensitive to include duration (of calls and of connection with cell towers), time of day (calls are made and cell towers are connected), and frequency of caller usage. In order to classify a legitimate user over a mobile network the thesis then validates this approach by implementing C4.5, RIPPER and SOM algorithms over MIT’s Reality Mining Dataset. The results support the proposed architecture and present accuracy rate as high as 96% for call logs and 94% for tower logs under training conditions.

With more than 5 billion users globally, mobile devices have become ubiquitous in our daily life. The modern mobile handheld device is capable of providing many multimedia services through a wide range of applications over multiple networks as well as on the handheld device itself. These services are predominantly driven by data, which is increasingly associated with sensitive information. Such a trend raises the security requirement for reliable and robust verification techniques of users.This thesis explores the end-user verification requirements of mobile devices and proposes a novel Behaviour Profiling security framework for mobile devices. The research starts with a critical review of existing mobile technologies, security threats and mechanisms, and highlights a broad range of weaknesses. Therefore, attention is given to biometric verification techniques which have the ability to offer better security. Despite a large number of biometric works carried out in the area of transparent authentication systems (TAS) and Intrusion Detection Systems (IDS), each have a set of weaknesses that fail to provide a comprehensive solution. They are either reliant upon a specific behaviour to enable the system to function or only capable of providing security for network based services. To this end, the behaviour profiling technique is identified as a potential candidate to provide high level security from both authentication and IDS aspects, operating in a continuous and transparent manner within the mobile host environment.This research examines the feasibility of a behaviour profiling technique through mobile users general applications usage, telephone, text message and multi-instance application usage with the best experimental results Equal Error Rates (EER) of 13.5%, 5.4%, 2.2% and 10% respectively. Based upon this information, a novel architecture of Behaviour Profiling on mobile devices is proposed. The framework is able to provide a robust, continuous and non-intrusive verification mechanism in standalone, TAS or IDS modes, regardless of device hardware configuration. The framework is able to utilise user behaviour to continuously evaluate the system security status of the device. With a high system security level, users are granted with instant access to sensitive services and data, while with lower system security levels, users are required to reassure their identity before accessing sensitive services.The core functions of the novel framework are validated through the implementation of a simulation system. A series of security scenarios are designed to demonstrate the effectiveness of the novel framework to verify legitimate and imposter activities. By employing the smoothing function of three applications, verification time of 3 minutes and a time period of 60 minutes of the degradation function, the Behaviour Profiling framework achieved the best performance with False Rejection Rate (FRR) rates of 7.57%, 77% and 11.24% for the normal, protected and overall applications respectively and with False Acceptance Rate (FAR) rates of 3.42%, 15.29% and 4.09% for their counterparts.

Keratoconus (MIM#148300), a common bilateral, progressive corneal thinning disorder, is the leading indication for corneal transplantation in the developed world. Keratoconus usually arises in the teenage years and presents a significant health burden in work-age adults. Despite the visual and social impact of keratoconus, our lack of understanding of the molecular pathology of keratoconus is a major obstacle to the development of new therapeutic approaches. This study represents the first reported application of massively parallel sequencing of mRNA (RNA-Seq) to perform whole genome transcriptome analysis of keratoconic keratocytes. Genes-enrichment analysis identified several pathway maps that are disrupted in keratoconic keratocytes and associated with the pathogenesis of keratoconus. Microarray gene expression was used to validate differentially expressed genes identified by RNA-Seq in a global manner, whereas quantitative reverse transcriptase polymerase chain reaction was performed on selected candidate genes. Wnt signalling, TGF-beta signalling. ECM-matrix interactions, oxidative stress and inflammatory- related genes were specifically identified in keratoconic keratocytes and implicated in the pathogenesis of keratoconus. Analysis of individual target genes identified altered expression of both known and novel keratoconus-related genes, in particular, SFRP1, BMP4, CBS, POSTN, C0L11Al, COL4Al, SOD1, IL6, and SP3. Functional analyses and expression profiling of keratoconic keratocytes harbouring a novel heterozygous missense mutation (c.l920G>T; p.Gln640His) in the zinc finger E-box binding gene 1 (ZEB1) was also performed. The mutant ZEB1 protein was stable and localised to the nucleus resulting in an enhanced transcriptional repressor of known ZEB1 targets, involved in epithelialmesenchymal transition and collagen synthesis. This ZEB1 mutation results in a gain-In-function with enhanced transcriptional repression of a number of gene targets associated with keratoconus, corneal thickness and Fuchs endothelial corneal dystrophy. This study has identified a number of molecular targets for keratoconus and provides a significant insight into the gene pathways involved in keratoconus pathogenesis. Further functional studies can build on this evidence to interrogate disease pathogenesis, identify novel genes and develop molecular therapies for keratoconus.

The different classes of leukaemia comprise numerous heterogeneous subgroups, which differ in their cellular and molecular characteristics. Accurate risk stratification is essential for tailoring of therapy. Based upon clinical features and cytogenetic and molecular diagnostics, the majority of ALL patients are assigned to one of the following prognostically significant subtypes having; ETV6-RUNX1, BCR-ABL1 or TCF3-PBX1 fusions, MLL rearrangements, high hyperdiploidy (HeH) with >50 chromosomes, hypodiploidy or T-ALL. Whilst these genetic abnormalities are important in leukaemogenesis, and provide diagnostic and prognostic markers, co-operating oncogenic aberrations are often required for the production of a full leukaemic phenotype. A number of additional aberrations have been identified within some cytogenetic subgroups, but the full complement of cooperating abnormalities, and their distribution within ALL subtypes remains to be defined. In this study the genomic changes in a total of 94 ALL patients from the ETV6-RUNX1 (n=34), iAMP21 (n=19) and unclassified (n=41) patient subgroups were characterised using cytogenetics, FISH, array comparative genomic hybridisation, molecular copy number counting, mutation analysis, qRT-PCR and targeted gene expression arrays. This approach allowed the identification of a range of large scale and submicroscopic aberrations, targeting multiple known and novel regions, including a number of genes. The elucidation of pathways and genes dysregulated in these malignant subgroups has provided further insight into the underlying cause of the disease phenotype. The PDE9A and TBL1XR1 genes were implicated in the pathogenesis of iAMP21 and ETV6-RUNX1 patients, respectively. Additionally, the potential prognostic significance of the ADD3 gene in ETV6-RUNX1 positive ALL was revealed.

Abstract This thesis has shown that the development of the instrumentation necessary to provide in-situ thermal imaging for the determination of homogeneity of concrete is theoretically sound. Drilled shafts are large diameter underground cast-in-place columns that necessarily rely on sound integrity to properly withstand imposed loadings. As a by-product of the most common construction techniques, the entire process is often completely blind whereby the excavation and concreting processes are conducted beneath the surface of the water table (or slurry level). This results in an inability to inspect the final product and in many cases allows anomalous inclusions (soil cave-ins, slurry pockets, etc) to go undetected especially when they are formed outside the steel reinforcing cage. In an effort to gain verification of the as-built, below ground structure, numerous non-destructive test methods have been devised. Each of these methods have merits and drawbacks with regards to the full extent of the tested concrete volume. To further this cause, a new methodology was developed that uses the energy from hydrating concrete to assess the presence or absence of an intact concrete. Therein, the temperature generated by the curing concrete can be measured and correlated to the probable dimensions of the drilled shaft. This thesis outlines the development of the instrumentation capable of making in-situ temperature measurement of drilled shafts to assure the homogeneity of concrete is acceptable. To that end, several configurations of instrumentation approach were tested on varying scales from small lab specimens to full-size field constructed drilled shafts. The bulk of this work was conducted several years before the completion of the thesis and has the benefit of noting later developments. For instance, this study was used to seed future research and led to subsequent FDOT and WSDOT (Washington State DOT) funded research for the express purpose of identifying capabilities of thermal testing in those states. Likewise, present day practice and use of the approach has also been documented.

Approved for public release; distribution is unlimited
For years it has been the goal of the Intelligence Community to limit sharply or even put a stop to terrorism, be it through the prevention of attacks or apprehension of those seeking to commit such acts. While there has been some success, perhaps further progress could be made by incorporating techniques used by law enforcement into the intelligence process. Geographic profiling has been used successfully by law enforcement agencies to aid in the capture of serial criminals, and due to the similarities between the two, it is possible that geographic profiling could do the same against terrorists. In the case of the Abu Sayyaf Group in the southern islands of Sulu and Basilan in the Philippines, geographic profiling techniques were partially successful in highlighting the possible future locations or types of incidents that would next be committed by the group. The success was limited, but while it may not be the next great breakthrough in the prevention of terrorist attacks, it appears to be another layer of analysis that can be incorporated in the intelligence cycle.

Approved for public release, distribution is unlimited
The Integrated Theater Assessment Profiling System (iTAPS) takes the original stove-piped Theater Assessment Profiling System (TAPS) software solution and turns it into a robust, data-centric, web-based decision support system for Commander, Second Fleet. ITAPS uses the .Net Framework and ASP.NET/ADO.NET, along with SQL Server to provide a web-enabled application that gives an overarching, abstracted view of the battle space for the Operational Commander while still providing drill-down capability and trend analysis tools if more detail is desired. The software was developed using the extreme programming technique and black box testing methods. A demonstration was performed at Second Fleet to test its acceptability and usability.

The pharmaceutical analysis of illicit drugs and the associated impurity profiling can contribute to a harm reduction approach. Impurities in illicit drugs represent a complex problem that requires detailed and reproducible analysis, in part because cutting agents are not often reported by the forensic science community. This research project is focussed on characterising Novel Psychoactive Substances (NPS) and determining their purity/impurity (adulteration) employing a wide range of spectroscopic and chromatographic methods, mainly NMR spectroscopy and Mass Spectrometry. Another aim is to develop a routine, rapid, and quantitative analytical protocol to identify illicit drugs and their impurities (cutting agents) in seized street samples. These two major aims were achieved. A comprehensive, but critical review of the current literature is provided with respect to the analytical chemistry aspects of illicit drugs and especially their cutting agents. This literature review has a focus on the global concern arising from the current and continuing emergence of NPS and their diverse public health effects. Evidence is provided of illegal drugs, mainly so-called “legal highs” (NPS), from detailed analysis of the contents of amnesty bins and their adulterants provided by the Police from a Bristol night club and from the 2013 Glastonbury music festival, besides other samples they had seized. An accurate chemical assignment of flephedrone regioisomers is made and compared with mephedrone. Impurity profiling of street mephedrone and ketamine samples and their adulterants is presented. A possible link between mephedrone samples is investigated by applying PCA and HCA statistical methods to the 1H NMR data.

Wastewater-based epidemiology (WBE) is an innovative tool that, unlike traditional epidemiological approaches, is capable of providing real-time profiling of community health and lifestyle along with emerging trends and changes in pattern usage of drugs in wastewater. By using human urinary excreted indicators, so-called “biomarkers”, WBE provides estimates at population level. Therefore, the choice and the evaluation of suitable biomarkers of exposure to drugs is of fundamental importance for the public health monitoring. Moreover, since many drugs are chiral, the investigation on enantiomeric profiling of chiral biomarkers provides a new dimension to WBE. To aid enantiomeric profiling in WBE, sensitive enantioselective methodologies are required. In this thesis, two novel multiresidue analytical methods based on chiral liquid chromatography coupled to mass spectrometry were developed and validated. The first method investigated the main human biomarkers for the detection of illicit drugs of abuse and potentially abused licit drugs. New biomarkers were investigated, such as mephedrone, PMA and all MDMA’s metabolites. Furthermore, a case study on mephedrone posed the basis for a novel approach towards biomarker selection in estimation of human exposure to chiral drugs with limited metabolism data. As a result, mephedrone was a suitable biomarker due to its stability in wastewater. In addition, some of its metabolites were also proposed as potential candidates for mephedrone use. The second method explored biomarkers of quinolones’ use, as they represent one class of antibiotics with rising concern in antibiotic resistance. The most comprehensive panel of quinolones’ biomarkers was considered for the first time in WBE studies. Both methodologies were applied to wastewaters from eight locations in Europe allowing the first pan-European studies on enantiomeric profiling of chiral biomarkers. Key findings of this research included: the detection of high mephedrone loads only in the UK, thus indicating human consumption; the detection of HMMA, a MDMA metabolite, as a suitable indicator of MDMA consumption and the determination of different synthetic production routes of methamphetamine across Europe. With regards to quinolones’ biomarkers, higher ofloxacin loads were found in Southern European cities along with differences in enantiomeric fraction with respect to Northern ones. Moreover, ofloxacin’s metabolites showed ofloxacin use and ulifloxacin was found as a result of prulifloxacin consumption. Therefore, enantiomeric profiling led to an understanding of: (i) new patterns of emerging drugs of abuse, (ii) changes in patterns of classical illicit drugs and (iii) quinolones with the verification of the origin of drug residue, potency of abused drug and its synthetic route and (iv) quinolones’ metabolic profiles. Moreover, the simultaneous determination of quinolones’ biomarkers in European samples allowed for the verification of spatial and temporal trends of quinolones’ use and the occurrence of their resistance genes. This proof-of-concept research will facilitate further advances in the WBE field.

This thesis has investigated the offence of robbery. Specifically, the semi-systematic review analysed commercial armed robbery, grouping offenders in terms of an apparent scale of professionalism to amateurism. Within armed robbery, target hardening strategies appear to have reduced opportunities for professionals, with a corresponding increase in amateur armed robbers fuelled by drug habits. The empirical study found that levels of interaction used by an offender with a victim increased with offender age. Interaction was lower for a robbery committed in an external location and for offenders with previous convictions for offences against the person and property. The violence facet could not be labelled as a specific discriminatory predictor. The findings from the research and semi-systematic review distinguished between two types of robbery offender; a career professional and an amateur antisocial robber. A career professional is older and more experienced, more likely to offend in a commercial location, commit the crime in a planned and controlled manner, use high levels of interaction and lower levels of violence. An amateur antisocial robber is more likely to commit an offence outside, have previous convictions for offences against the person and property and/or be under the influence of an illegal substance. The offence is likely to be opportunistic and chaotic, characterised by high levels of violence and low levels of interaction. The Inventory of Offender Risk, Needs and Strengths (IORNS) psychometric measure was analysed. It has the potential to provide an assessment of a robbery offender‟s ongoing treatment and risk management. However, it requires further validation and reliability analysis before it is deemed appropriate in doing so. The case study highlighted the impact of cannabis misuse on a robbery offender‟s behaviour pattern and mental illness. Implications for offender treatment needs, future therapeutic intervention and risk management are discussed along with the need for further validation of the proposed model.

Cisplatin is a first line chemotherapeutic agent for lung cancer however, although patients may respond to therapy, resistance often develops with tumour recurrence and disease progression. Somatic alterations in the tumour may alter therapeutic responses. Consequently a model of cisplatin resistance in lung cancer derived A549 cells has been created to examine the genomic changes that occur as chemo-resistance develops. Drug resistance was induced in A549 cells through multiple rounds of cisplatin dosage and recovery over two different time courses. The concentration of cisplatin required to inhibit growth (inhibitory concentration [IC] value) was calculated at each round and cycles were continued until the IC value increased at least four-fold. Cells were harvested and total RNA extracted for whole transcriptome microarray analysis. Data was analysed using R statistics and associated packages, Affy, Limma, Mfuzz and WGCNA. A five-fold increase in IC value was generated over successive doses in both regimes, accompanied by highly significant changes in gene expression. To explore these changes, temporal expression clustering and extensive network analyses were performed across the rounds of cisplatin dosing, as well as an untreated cell culture time course that acted as a comparison to the two treated regimes. The results gathered from this robust model suggest that differences in dose and frequency of chemotherapy may affect genomic changes at specific loci that confer cisplatin resistance. Interesting and relevant pathways and genes have been discovered. In combination with analyses on a small patient cohort, these results have provided insights into the mechanism of cisplatin resistance and have highlighted new clinical biomarkers of potential use in prognosis of patients undergoing cancer treatment.

Thesis (PhD)--Stellenbosch University, 2014.
ENGLISH ABSTRACT: Soft, serene insect songs add an intrinsic aesthetic value to the landscape. Yet these songs also have an important biological relevance. Acoustic signals across the landscape carry a multitude of localized information allowing organisms to communicate invisibly within their environment. Ensifera are cryptic participants of nocturnal soundscapes, contributing to ambient acoustics through their diverse range of proclamation songs. Although not without inherent risks and constraints, the single most important function of signalling is sexual advertising and pair formation. In order for acoustic communication to be effective, signals must maintain their encoded information so as to lead to positive phonotaxis in the receiver towards the emitter. In any given environment, communication is constrained by various local abiotic and biotic factors, resulting in Ensifera utilizing acoustic niches, shifting species songs spectrally, spatially and temporally for their optimal propagation in the environment. Besides the importance of Ensifera songs from an ethological point of view, the multitude of species-specific signals provide an acoustic tapestry representing species diversity across ecological gradients and over time. Acoustic inventorying and monitoring of the landscape can reflect the environmental status of ecological systems, from natural to disturbed by human influence. In contrast to traditional survey techniques, sound recording and interpretation is a non-invasive method that allows for the detection and classification of highly cryptic, yet insightful indicators of ecosystem change. Here, acoustic monitoring was used across diverse ecological gradients to improve understanding of species diversity patterns, and how they change in response to both natural gradients and in response to the human dominated landscape mosaic. This study was undertaken in three geographic locations from tropical rain forest of Brunei, Borneo, to the landscapemosaic of plantation forestry in KwaZulu-Natal, South Africa, and to the botanically rich, mountain fynbos region of the Cape Floristic Region, also in South Africa. Each region provided a diverse and particular landscape to test the value of acoustic surveys for determining local diversity patterns across natural gradients and to assess the value of the technique for assessing the impact human influence across landscapes. In tropical rainforests, an entire acoustic guild was investigated to determine how acoustic species partition their acoustic communication channels spectrally, temporally and spatially, to avoid acoustic interference. The overall assemblage showed considerable spectral partitioning. Diurnally active species showed low temporal niche overlap, whereas nocturnal species did not utilize temporal partitioning. Lack of nocturnal temporal partitioning suggests other mechanisms of acoustic avoidance are sufficient to avoid acoustic overlap, or that there are insufficient cues to partition nocturnal acoustic environments. Acoustic species also utilized spatial partitioning, with distinct acoustic assemblages at vertical heights and with elevation. Utilization of a range of different strategies allow many species to communicate with conspecifics with little or no interference from other species in a signal rich environment. Acoustic profiling was also undertaken in KwaZulu-Natal, South Africa, across a plantation forestry landscape mosaic with diverse ecological gradients containing both alien and indigenous vegetation, as well as boarding large natural protected areas. Areas covered in alien timber or non-endemic grass were devoid of acoustic signals. Managed areas that were mowed and heavily grazed were not effective in maintaining the natural complement of nocturnal acoustic species. Within natural vegetation patches inside plantations, acoustic species richness increased with plant heterogeneity and patch size. Patches of indigenous vegetation within the plantation matrix effectively reduced the contrast of transformed landscapes with surrounding natural areas, with indigenous forest patches containing a highly characteristic acoustic species assemblage. Within the botanically rich, mountain fynbos region of the Cape Floristic Region, acoustic profiling was conducted across gradients of elevation, season and vegetation. Across these gradients, katydid acoustic signals were identified and characterized for the first time. This resulted in the discovery of two new katydid species and a novel sound producing structure in a carabid beetle, a species previously unknown to produce sound. Acoustic diversity across seasonal and elevational gradients increased with increasing temperatures. Climatic variability along the elevational gradient produced variation in seasonal phenology. Katydids also utilized high frequency acoustic signals, which is probably an adaptation to overcome background noise from wind, so prevalent in this area. Furthermore, despite producing conspicuous signals for mate attraction and pair formation, katydids were found not to be part of bat-eared fox diet, an insectivorous, nocturnal predator that uses its characteristic large ears to detect sounds made by invertebrate prey. This study shows the value of using acoustic emissions from katydids to identify acoustic diversity patterns across ecological gradients and in response to human impacts on the landscape.
AFRIKAANSE OPSOMMING: Sagte, rustige insekliedjies voeg ‘n intrinsieke estetiese waarde aan die landskap. Tog is hierdie liedjies ook van belangrike biologiese waarde. Akoestiese seine oor die landskap dra ‘n magdom plaaslike inligting wat organismes in staat stel om onsigbaar te kommunikeer binne hul omgewing. Langhoringsprinkane is kriptiese deelnemers van die nagtelike klankomgewing en dra by tot die omringende akoestiek deur hul verskeidenheid van proklamasieliedjies. Alhoewel dit nie sonder inherente risiko’s en beperkings is nie, is die belangrikste funksie van seine seksuele advertering en paarvorming. Vir akoestiese kommunikasie om effektief te wees, moet seine hul geënkodeerde inligting handhaaf, sodat dit sal lei tot positiewe fonotaksis in die ontvanger teenoor die emittor. In enige gegewe omgewing, word kommunikasie beperk deur verskeie plaaslike abiotiese en biotiese faktore. Dit lei tot die gebruik van akoestiese nisse deur langhoringsprinkane, wat hulle liedjies spektraal, ruimtelik en temporeel aanpas vir optimale verspreiding in die omgewing. Benewens die belang van die langhoringsprinkaan liedjies uit ʼn etologiese oogpunt, bied die menigte spesie-spesifieke seine ʼn akoestiese tapisserie wat spesiesdiversiteit verteenwoordig oor ekologiese gradiënte en oor tyd. Akoestiese opname en monitering van die landskap kan die omgewingstoestand van ekologiese stelsels weerspieël, van natuurlike tot menslik versteurde stelsels. In teenstelling met tradisionele opnametegnieke, is klankopname en interpretasie ʼn nie-indringende metode wat dit moontlik maak om hoogs kriptiese, nog insiggewende indikators van ekosisteemverandering op te spoor en te klassifiseer. In hierdie studie is akoestiese monitering gebruik oor diverse ekologiese gradiënte om ons begrip te verbeter van spesies diversiteitspatrone, en hoe dit verander in reaksie op beide natuurlike gradiënte en in reaksie op die menslik gedomineerde landskapmosaïek. Hierdie studie is onderneem in drie geografiese liggings: tropiese reënwoud in Brunei, Borneo, die landskapmosaïek van plantasiebosbou in KwaZulu-Natal, Suid-Afrika, en die plantryke, bergfynbos-streek van die Kaap Floristiese Streek, ook in Suid-Afrika. Elke streek het 'n diverse en besondere landskap verskaf om die waarde van akoestiese opnames te toets vir die bepaling van plaaslike diversiteitspatrone in natuurlike gradiënte, asook om die waarde van die tegniek te bepaal vir die beoordeling van die impak van menslike invloed oor landskappe. In tropiese reënwoude, is 'n hele akoestiese gilde ondersoek om te bepaal hoe akoestiese spesies hul akoestiese kommunikasiekanale spektraal, ruimtelik en temporeel verdeel om akoestiese inmenging te vermy. Die algehele groep het aansienlike spektrale verdeling getoon. Dagaktiewe spesies het lae temporele nisoorvleueling getoon, terwyl nagtelike spesie nie temporele verdeling benut het nie. Gebrek aan nagtelike temporele verdeling dui daarop dat ander meganismes van akoestiese vermyding voldoende is om akoestiese oorvleueling te vermy, of dat daar onvoldoende seine is om nagtelike akoestiese omgewings te verdeel. Akoestiese spesies het ook ruimtelike verdeling benut, met verskillende akoestiese spesiesversamelings op vertikale hoogtes en met hoogte bo seevlak. Die gebruik van 'n verskeidenheid van strategieë maak dit moontlik vir spesies om te kommunikeer met min of geen inmenging van ander spesies in 'n seinryke omgewing. Akoestiese profielsamestelling is ook onderneem in KwaZulu-Natal, Suid-Afrika, oor 'n plantasiebosbou landskapmosaïek met diverse ekologiese gradiënte wat beide uitheemse en inheemse plantegroei, sowel as groot, natuurlike, beskermde gebiede ingesluit het. Gebiede wat bestaan het uit uitheemse timmerhoutbome of nie-endemiese gras, was heeltemal sonder akoestiese seine. Bestuursgebiede wat gesny en swaar bewei was, het nie doeltreffend die natuurlike komplement van nagtelike akoestiese spesies gehandhaaf nie. In natuurlike plantegroei fragmente binne plantasies, het akoestiese spesiesrykheid toegeneem met plantverskeidenheid en fragmentgrootte. Fragmente van inheemse plantegroei binne die plantasiematriks het effektief die kontras van getransformeerde landskappe met omliggende natuurlike gebiede verminder, en inheemse woudefragmente het hoogs kenmerkende akoestiese spesiesversamelings gehad. Binne die plantryke bergfynbosstreek van die Kaap Floristiese Streek, was akoestiese profielsamestelling gedoen oor gradiënte van hoogte bo seevlak, seisoen en plantegroei. Oor hierdie gradiënte, is langhoringsprinkaan akoestiese seine geïdentifiseer en gekenmerk vir die eerste keer. Dit het gelei tot die ontdekking van twee nuwe langhoringsprinkaan spesies en 'n nuwe klankvervaardiging struktuur in 'n Carabid kewer, 'n spesie wat nie voorheen bekend was om klank te produseer nie. Akoestiese diversiteit oor gradiënte van seisoen en hoogte bo seevlak het toegeneem met toenemende temperature. Variasie in klimaatstoestande oor die gradiënt van hoogte bo seevlak het variasie in seisoenale fenologie veroorsaak. Langhoringsprinkane het ook hoë frekwensie akoestiese seine benut, wat waarskynlik 'n aanpassing is om agtergrondgeraas van die wind, wat so algemeen is in hierdie gebied, te bowe te kom. Verder, ten spyte van die vervaardiging van opvallende seine vir maataantrekking en paarvorming, het langhoringsprinkane nie deel gevorm van die bakoorjakkals se dieet nie. Hierdie is 'n insekvretende, nagtelike roofdier wat gebruik maak van sy kenmerkende groot ore om klanke op te spoor wat gemaak word deur invertebraatprooi. Hierdie studie toon die waarde van die gebruik van akoestiese seine van langhoringsprinkane om akoestiese diversiteitspatrone te identifiseer oor ekologiese gradiënte en in reaksie op menslike impakte op die landskap.

The process of embryogenesis in higher plants is a critical stage of the sporophytic life cycle, transforming the fertilised egg cell via a precise sequence of events into a multi-cellular organism. It is during embryogenesis that the body plan of the developing plant is established. Analysis of transcriptional changesembryo. This thesis demonstrates the application of laser capture micro dissection to the analysis of embryogenesis in the model plant Arabidopsis thaliana. This technique has been used in combination with DNA microarray technology to allow a global analysis of gene expression in the cotyledon, root and shoot apical meristem regions of the torpedo-stage embryo. Validation of the approach has been achieved by comparison of the ATH1 GeneChip® data obtained, with published gene expression patterns confirmed by in situ hybridisation and promoter: GUS analysis. Further validation was successfully undertaken through the creation of promoter: Gus constructs for a number of previously uncharacterised putative transcription factor genes, selected on the basis of differential expression between the cotyledon and root regions. Initial attempts to assign putative function to these genes through an analysis of T-DNA insertion lines yielded no aberrant phenotypes. Transcriptional profiling of embryogenesis from the globular-stage through to the torpedo-stage was carried out using GeneSpring, uncovering distinct spatial and temporal expression patterns, and revealing a number of genes of potential interest for further research.

Context In the present world, there is an increase in the usage of the telecommunication networking services, as there is a need of efficient networking services in various fields which can be obtained by using the efficient networking components. For that purpose we have to know about the components parameters. One of the most important parameter is the energy usage of networking components. Therefore, there is a need in power profiling of the network switches. Objectives The objective of this research is to profile the power usage of different network components(Switches) for various load scenarios. Power measurements are done by using the open energy monitoring tool called emonpi. Methods The research method has been carried out by using an experimental test bed. In this research, we are going to conduct the experiments with different configurations to obtain different load conditions for sources and destinations which will be passed through DUT(Device Under Test). For that DUT’s we will measure power usage by monitoring tool called emonpi. Then the experiments are conducted for different load scenarios for different switches and results are discussed. Conclusion From the results obtained, the Power profiles of different DUT’s are tabulated and analyzed. These were done under different ports and load scenarios for Cisco2950, Cisco3560 and Netgear GS-724T. From the results and analysis it can be stated that the power usage of Cisco 2950 is having the maximum power usage in all the considered scenarios with respect to packet rate and also number of active ports. The Netgear-GS724T is having the minimum power usage from the three switches as it having the green switch characteristics in all scenarios. And the Cisco 3560 is in between the above two switches as it is having energy efficient management from Cisco. From this we have proposed a simple model for energy/power measurement.

Mestrado em Finanças
Retail banks classification of clients according to their risk profile has been a theme very much under discussion in the last few years. In this paper we aim to investigate, for the affluent and private banking client segment, if the perception that people have of their risk profile effectively corresponds to their real risk profile as given by a measure of historical volatility. In addition, we analyze, on the real portfolio of financial investments, the impact of each of the following factors in regard to people's attitude towards risk: age, gender, degree of education, financial situation and investment experience.

The goal of the Instant Knowledge project was to design a system to facilitate the sharing of knowledge and expertise within a distributed mobile environment. This system automatically builds profiles of experts interests, and automatically recommends them based on context and social networking information. This thesis describes my contributions to the IK project which involves profiling users and making recommendations using machine learning techniques. Recommender systems are information filtering systems which recommend items to users based on a model of their preferences. Recommenders suffer from a number of problems: they do not make use of contextual information, so recommendations may be untimely or inappropriate; they often use a centralised architecture, which makes it diffcult to react to the changing needs of users; they are often implemented in an ad-hoc fashion making it difficult to make principled improvements or add extra information. In this thesis I present a probabilistic recommender based on Bayes' theorem. Rating behaviour is modelled using a Bayesian prior to improve performance in conditions of data sparsity. The best results are obtained using a Gaussian model for user ratings, and a Gaussian-gamma model for co-rating behaviour. The use of a probabilistic framework should make it easier to add context information to the recommendation process. Generating profiles automatically carries the risk of accidentally including private information which may be discovered by querying the Instant Knowledge system. This presents a privacy risk, as private information may be accidentally incorporated into experts' profiles. I present a framework for evaluating the effect of contamination on performance, and the ability of filtering techniques to preserve privacy. Several filtering techniques are tested and I show that supervised and semi-supervised naive Bayes classifiers can help to preserve privacy.

Sequencing technologies and applications have pushed the limits and enabled novel studies of biological mechanisms, evolutionary relationships and communication networks between cells. The technical developments leading to single cell RNA-sequencing have enabled detection of rare cell populations while spatial resolution added insights into larger biological environments, like tissues and organs. Massively parallel sequencing has paved the way for integrated high-throughput analyses including that of studying gene expression, protein expression and mapping of microbial communities. This thesis starts with an introduction describing the technical and biological advancements made in recent years with focus on spatially resolved approaches. Then, a summary of recent accomplishments is presented, which enabled ongoing work in a novel field of spatial hostmicrobiome profiling. Lastly, the concluding remarks include both a future perspective and a short reflection on the current developments in the spatial multi-omics field. 16S sequencing is often used for taxonomic classification of bacteria. In Paper I, this sequencing technique was used to study the aerodigestive microbiome in pediatric lung transplant recipients. Many of these patients regretfully reject the organ after transplant, but the underlying cause is, in many cases, unknown. In this paper, multiple factors influencing rejection were examined including that of the aerodigestive microbiome. Pediatric lung transplant recipients often suffer from gastrointestinal dysmotility and the focus of this study was also to analyze changes in the microbiome in relation to irregular gastric muscle movements. The results showed that lung transplant recipients had, in general, lower microbial diversity in the gastric fluid and throat and also that the microbial overlap between lung and gastric sampling sites was significantly less in transplant recipients compared to controls. In addition, gastrointestinal dysmotility was shown to influence the gastric microbiome in lung transplant recipients, but, given the small sample size available in this study, the correlation to patient outcome could not be examined. Integrated analysis of the transcriptome and the antibody-based proteome in the same tissue section was enabled using the method developed in Paper II. Spatial Multi- Omics (SM-Omics) uses a barcoded glass array to capture mRNA and antibody-based expression of selected proteins in the same section. The antibody-based profiling of the tissue section was enabled by either immunofluorescence or DNA-barcoded antibodies that were then decoded by sequencing. The protocol was scaled-up using an automated liquidhandling system. Using this method, simultaneous profiling of the transcriptome and multiplexed protein values was determined in both the mouse brain cortex and mouse spleen. Results showed a high correlation in spatial pattern between gene expression and antibody measurements, independently of the antibody labelling technique. SM-Omics generates a high-plex multi-omics characterization of the tissue in a high throughput manner while exhibiting low technical variation.
Tekniker och applikationer som använder sekvensering har flyttat fram gränsernaoch tillåtit nya undersökningar av biologiska mekanismer, evolutionära släktskap ochkommunikationsnätverk mellan celler. De tekniska utvecklingarna som har lett fram tillRNA-sekvensering av enskilda celler har möjliggjort upptäckten av sällsynta cellpopulationer medan den rumsliga upplösningen har inneburit en ökad förståelse av störrebiologiska miljöer, såsom vävnader och organ. Massively parallel sequencing har banat vägför integrerade analyser med hög kapacitet, vilket inkluderar analys av genuttryck,proteinuttryck och kartläggning av bakteriella samhällen. Den här avhandlingen börjar meden introduktion som beskriver tekniska och biologiska framsteg som gjorts de senaste åren,med fokus på den rumsliga upplösningen. Sedan följer en summering av de senasteprestationerna som har möjliggjort det pågående arbetet i ett nytt fält som avhandlarrumslig profilering av bakterien och dess värd. Slutligen innehåller slutordet både ettframtida perspektiv samt en kort reflektion av den nuvarande utvecklingen inom fälten förrumslig mång-omik. 16S-sekvensering används ofta för att taxonomiskt klassificera bakterier. Dennasekvenseringsteknik användes i artikel I för att studera mikrobiomet i luft- ochmatspjälkningskanalen hos barn med transplanterad lunga. Dessvärre är det vanligt medavstötning av lungan efter transplantationen hos många av dessa patienter, men denunderliggande orsaken till avstötningen är, i många fall, okänd. I denna studie undersöktesflertalet faktorer, inklusive mikrobiomet i luft- och matspjälkningskanalen, som kan tänkaspåverka bortstötningen. Barn med transplanterad lunga lider ofta av störningar i magtarmkanalens rörelser och artikelns fokus var därmed även att analysera förändringar imikrobiomet i relation till dessa avvikande rörelser i mag-tarmkanalen. Resultatet visade attpatienter med transplanterad lunga generellt hade lägre bakteriell mångfald i magsaft ochhals, samt att det bakteriella överlappet mellan lunga och magsaft var signifikant mindre ipatienter med transplanterad lunga jämfört med kontrollerna. För övrigt visade det sig attstörningar i mag-tarmkanalens rörelser påverkade magsaftens mikrobiom hos patientermed transplanterad lunga, men på grund av studiens storlek på urvalet, kunde det inteundersökas hur detta korrelerade till utfallet hos patienterna. Integrerad analys av transkriptomet och antikroppsbaserad analys av proteomet isamma vävnadssnitt har möjliggjorts genom metoden som utvecklats i artikel II. SpatialMulti-Omics (SM-Omics) använder ett avkodningsbart mönster av korta DNA-segment påen glasyta för att fånga mRNA och antikroppsbaserat uttryck av utvalda proteiner frånsamma vävnadssnitt. Den antikroppsbaserade profileringen av vävnadssnittet uppnåddesgenom antingen immunofluorescens eller antikroppar märkta med DNA-segment somkunde avkodas genom sekvensering. Protokollet skalades upp genom ett automatiseratsystem för att behandla vätskor. Genom användning av denna metod kunde simultanprofilering av transkriptomet och flertalet proteiner uppnås i både hjärnbarken och mjältenhos en mus. Resultaten visade en hög korrelation i det rumsliga mönstret mellangenuttrycket och de antikroppsbaserade mätningarna, oberoende av hur antikropparnahade märkts. SM-Omics genererar en storskalig karaktärisering av vävnaden av flera omikermed hög kapacitet samtidigt som den har låg teknisk variation.

QC 2021-02-02

Complex codebases with several layers of abstractions have abundant inefficiencies that affect the performance. These inefficiencies arise due to various causes such as developers' inattention to performance, inappropriate choice of algorithms and inefficient code generation among others. To eliminate the redundancies, lots of work has been done during the compiling phase. However, not all redundancies can be easily detected or eliminated with compiler optimization passes due to aliasing, limited optimization scopes, and insensitivity to input and execution contexts act as severe deterrents to static program analysis. There are also profiling tools which can reveal how resources are used. However, they can hard to distinguish whether the resource is worth fully used. More profiling tools are in needed to diagnose resource wastage and pinpoint inefficiencies. We have developed three tools to pinpoint different types of inefficiencies in different granularity. We build Runtime Value Numbering (RVN), a dynamic fine-grained profiler to pinpoint and quantify redundant computations in an execution. It is based on the classical value numbering technique but works at runtime instead of compile-time. We developed RedSpy, a fine-grained profiler to pinpoint and quantify value redundancies in program executions. Value redundancy may happen overtime at the same locations or in adjacent locations, and thus it has temporal and spatial locality. RedSpy identifies both temporal and spatial value locality. Furthermore, RedSpy is capable of identifying values that are approximately the same, enabling optimization opportunities in HPC codes that often use floating-point computations. RVN and RedSpy are both instrumentation based tools. They provide comprehensive result while introducing high space and time overhead. Our lightweight framework, Witch, samples consecutive accesses to the same memory location by exploiting two ubiquitous hardware features: the performance monitoring units (PMU) and debug registers. Witch performs no instrumentation. Hence, witchcraft - tools built atop Witch - can detect a variety of software inefficiencies while introducing negligible slowdown and insignificant memory consumption and yet maintaining accuracy comparable to exhaustive instrumentation tools. Witch allowed us to scale our analysis to a large number of codebases. All the tools work on fully optimized binary executable and provide insightful optimization guidance by apportioning redundancies to their provenance - source lines and full calling contexts. We apply RVN, RedSpy, and Witch on programs that were optimization targets for decades and guided by the tools, we were able to eliminate redundancies that resulted in significant speedups.

Measuring the use of resources in a production process has been a subject under great scrutiny since more than a hundred years ago. Traditionally, costing systems and cost accounting systems have been in charge of such functions in manufacturing corporations. On the other hand, in recent years Lean Manufacturing has become a powerful and popular force for change. A premier tool for process visualization and understanding is Value Stream Mapping, and it focuses primarily in the time dimension of the processes. However, it is clear that the interaction of cost and time is very important. This is felt in everyday occurrences, such as paying interests for credit cards, mortgages and other types of loans. It is intuitive that the longer a certain amount of money is held, the more it costs. Also, if a larger amount of money is held for one day, it will obviously cost more than holding a smaller amount of money. Therefore, cost and time, BOTH, determine the real cost of the use of money. However, this simple perception has not been applied equally to the measurement of manufacturing processes. They usually concentrate on either cost or time, but seldom in both at the same time and their interaction. The Westinghouse corporation formalized the concepts of the Cost-Time Profile in 1993, based on work done there during several decades. Simply put, the Cost-Time Profile measures how much money is invested in the manufacturing process of a product and for how long, creating a chart that presents the accumulated cost at every point in time (Cost-Time Profile) and measuring the area under this curve (Cost-Time Investment), and then using this quantification to measure the bottom line impact. This research has accomplished two main things: the detailed consideration of the Cost-Time Profile (CTP) and the issues and factors that affect it, and the extension of the concepts to the new reality of Extended Enterprises. In a logical sequence, the basic concepts of CTP are defined and presented. Then, the extension of them to Inter-Enterprise environments follows. Successive sections present how to build a CTP and the Inter-Enterprise Cost-Time Profile (IE-CTP), as well as discussing the factors that should be taken into account to bring the IE-CTP to practical applications, such as the effect of batching; the interaction with existing accounting systems; the consideration of direct cost, overhead and profit and the relationships between companies in supply networks to build IE-CTPs. Then the issue of how to improve the results of the Cost-Time Investment (CTI) and CTP is addressed, and schedule optimization models are developed; generic improvement scenarios and lean implementation scenarios are discussed; some simulation studies are presented for cases when this tool has advantages over deterministic tools and an IE-CTP specific software tool is presented. After learning how to improve the CTP and CTI, a discussion about how to use it and implement it is presented, and finally the summary and conclusions close this research report, identifying the contributions presented and leaving open avenues for future research.
Ph. D.

Obesity is a worldwide health issue that has reached epidemic proportions, and is defined as increased white adipose tissue mass. This increase in adiposity is caused by either hypertrophy of existing mature adipocytes, or hyperplasia of pre-adipocytes, leading to increased adipocyte numbers. The current study used the murine 3T3-L1 pre-adipocyte cell line to explore in vitro the differentiation process from pre-adipocytes to mature adipocytes, termed adipogenesis. Lipidomic analyses demonstrated a shift in the predominant lipid species present; from phospholipids in the pre-adipocytes, to triglycerides in the mature adipocytes. This was expected from the morphological changes known to occur in this cell line, from fibroblastic pre-adipocytes, to spherical lipid-loaded mature adipocytes. The production of various eicosanoids was also investigated, and their concentration was greatest during the pre-adipocyte stage. This profile was also seen with arachidonic acid, a precursor in eicosanoid synthesis. These changes in lipid metabolism and eicosanoid production appeared to be linked, allowing the differentiation process and lipid accumulation to continue. The obese state is also associated with a chronic low-grade inflammation, and so the effects of TNF-α and IL-6 intervention on adipocyte metabolism were investigated. Differences in lipid mobilisation caused by these pro-inflammatory agents were suggested due to increases or decreases observed in the concentrations of various triglyceride and fatty acid species. Increases were observed in the concentration of various detected eicosanoid species from the arachidonic acid cascade, mainly prostanoid species. Effects of the anti-inflammatory agent dexamethasone were also investigated in mature 3T3-L1 adipocytes. It was associated with increases in the concentration of both triglyceride and fatty acid species, suggesting possible increase in lipogenesis and/or decrease in lipolysis. Increases in the production of various eicosanoid species from the arachidonic acid pathway were also observed. The majority of these species are pro-inflammatory; however, PGE2 is known to have both pro-and anti-inflammatory effects, and this may help to explain these findings. In conclusion, the work presented in this thesis has revealed how adipocyte metabolism changes in the naturally occurring stages of adipogenesis, as well as in response to pro- and anti-inflammatory intervention. Associations were observed between adipokine gene expression, lipid metabolism and eicosanoid production; however, further work is required to confirm these links by identifying the underlying mechanisms involved.

Keloid disease (KD) is a benign fibroproliferative dermal disease of unknown aetiopathogenesis that occurs in genetically susceptible individuals. KD shows high heterogeneity within the lesion, harbouring different immune cell profiles, which are poorly characterised in KD at different lesional sites. Although, it has long been appreciated that chronic inflammation and dermal fibrosis is associated with other fibrotic diseases (e.g. scleroderma), this link has not, yet, been established in KD through direct evidence. Additionally, the limited availability of a simple KD animal model has hindered our understanding of the underlying pathogenesis of KD. Therefore, the main objectives were a) to identify and profile different immune cells at defined KD lesional and histological sites, b) to further characterize the potential contribution of viral particles in KD by investigating the gene and protein expression profile of toll like receptors that recognise viral particles in KD, and c) to develop an optimized long-term serum-free organ culture (OC) model for KD research as a tool for probing novel hypotheses in KD pathobiology deduced from a) and b) and to also validate the reliability and instructiveness of this novel ex vivo KD model with conventional (e.g. dexamethasone) and potential future anti-KD compounds [(-)-epigallocatechin-3-gallate (EGCG) and plasminogen activator inhibitor-1 (PAI-1) knock-down by siRNA]. To achieve above objectives, different cellular and molecular techniques were applied. Immune profiling of KD (chapter 2) at defined lesional and histological sites generated the first comprehensive analysis of KD-associated inflammatory cell infiltrates. This work demonstrated for the first time the presence of specific type of chronic inflammation in KD that resembles the formation of tertiary lymphoid tissues (TLTs) (in 14.7%, out of 68 KD cases). Although, these TLTs are not strictly linked to defined lesional sites within the KD, they are similar in structure to mucosa-associated lymphoid tissue (MALT). Therefore, we named this phenomenon as keloid-associated lymphoid tissue (KALT). Immunophenotyping of KD lesional sites also showed a predominance of T-cells, B-cells, M2 macrophages and OX40L+ degranulated mast cells in intralesional and perilesional sites of KD compared to normal skin and normal scar tissue. In the epidermis, Langerhans cells showed no changes, whereas the intra-epidermal T-cells were significantly increased in both the intralesional and perilesional sites of KD with an increased CD4:CD8 ratio. Intra-epidermal B-cells were only rarely found in KD. Interestingly, there was no significant statistical difference between intralesional and perilesional sites of KD immunophenotyping. These abnormal immune profiles suggest the persistence of non-resolving inflammation presence towards unknown stimuli, which require further investigation. The chronic inflammation could be followed by a reparative phase in a repetitive manner leading to KD formation. Evaluation of toll-like receptor (TLR) gene and protein expression in KD showed a significant increase in the expression of intra-epidermal TLR-6, -7 and dermal TLR-8. Since these TLRs are typically up regulated during anti-viral responses, these results further support the hypothesis that certain viruses or yet unidentified ligand may play a role in KD pathogenesis (chapter 3). A successful long-term, serum-free keloid OC model was established using a 4 mm sized punch biopsy embedded in collage matrix as air liquid interface in supplemented William’s E medium for up to 6 weeks (Chapter 4).

Background and aims Early detection of oesophageal cancer is the most effective strategy to improve its outcome. The thesis explores the metabolic characteristics of oesophageal tumours as compared to normal tissue from same patients as well as that from healthy individuals. The aim is to characterise the basic metabolic profile of oesophageal cancer, detect possible evidence of field cancerization and to characterise effect of chemotherapy on oesophageal cancer tissue. The clinical benefit is to develop a non-invasive early test to be translated to the bed side/laboratory practice for screening, diagnosis and prediction of response to chemotherapy for patients with oesophageal tumours. Methods Using 1H Magic Angle Spinning – Nuclear Magnetic Resonance (MAS-NMR) spectroscopy of intact tissue, we generated metabolic profiles of tumour tissue, proximal histologically normal mucosa from cancer patients (PHINOM) and mucosa from a control group. Using multivariate regression and receiver-operator characteristic analysis we were able to identify a panel of metabolites discriminating malignant and histologically-normal tissues from cancer patients from that of controls. Results While 26% and 12% of the spectral profile regions were uniquely discriminating tumour or control tissue respectively, 5% of the profile exhibited a significant progressive change in signal intensity from controls to PHINOM to tumour. Regions identified were assigned to phosphocholine, glutamate, myo-inositol, adenosine-containing compounds, uridine4 containing compounds and inosine. In particular, the phosphocholine to glutamate metabolite ratio in histologically-normal tissue signified the presence of oesophageal cancer (n=123, AUC 0.84, p<0.001). Metabolites that were found to be significantly altered in correlation to response to chemotherapy are lysine, glutathione, creatine, taurine, uracil and tyrosine Conclusion The results of the study enable better understanding of the metabolic perturbations in oesophageal and cancer. The findings support the hypothesis that field effects are present in oesophageal cancer, even in the absence of Barrett’s oesophagus. This indicates that metabolic profiling of tissue can potentially play a role in the surveillance of cancer especially in high risk groups. It also shows the potential of using this methodology to study the prediction of response to chemotherapy. This helps aiding the clinical decision making process and the implementation of personalized treatment protocols for different patients according to their disease extent and potential for response to chemotherapeutics.