Search

We develop new classes of semiparametric multivariate time series models based on Multi-Output Gaussian Processes and warped Multi-Output Gaussian Processes. These describe relationships between a current vector of observations and the lagged history of each marginal time series. We encode a serial dependence structure through mean and covariance functions and introduce a more complex dependence structure using copulas to couple each warped marginal Gaussian process. Within this class of models our primary goal is to detect causality and to study the interplay between the causal structure and the dependence structure. We do not, however, require true representation of the data generating process, but we model structural hypotheses regarding how causality may have manifested in the observed vector valued processes. With our framework we test the dependence with regards to the structures that are specified, and can use testing for causality under different model assumptions as a way to explore the data and the potentially complex dependence relationships. To perform the testing we consider several families of causality testing and develop compound tests which first require estimation/calibration of the mean and covariance functions parametrising the nonparametric vector valued time series. Our approach allows very general nonlinear dependence and causal relationships which are not often considered in classical parametric time series models, including causality in higher order information and joint extreme dependence features. We provide a generic framework which can be applied to a variety of different problem classes and discuss a number of examples to illustrate the ideas developed. We wish to detect the presence or absence of statistical causality where such extremal dependence features may or may not obfuscate the ability to detect causality in nonlinear partially observed time series models. The rationale for developing a semiparametric solution for modelling the partially observed time series is that we may accommodate, through the use of Gaussian Process models, a wide variety of features for the hypotheses about the trends and volatility and importantly their possible causal structures, which can be formally tested in our framework. Furthermore the use of Warped Gaussian Process models allows to incorporate higher order dependence such as extremal tail dependence features. Statistical Causality. The notion of causality that lies at the centre of our research is the concept of statistical causality, based on comparing two predictive models. Quoting Wiener [1956]: "For two simultaneously measured signals, if we can predict the first signal better by using the past information from the second one than by using the information without it, then we call the second signal causal to the first one". The type of casual dependence that is described by statistical causality is a mechanism that occurs at multiple lags over time - which could have been triggered by a sequence of processes, not an individual one. It can help to gain an insight into both cross-sectional and temporal dynamics of the data analysed. Warped Multi-Output Gaussian Processes. A Gaussian process is a Markov process, such that all finite dimensional distributions are Gaussian. While Gaussian processes models can accommodate wide range of properties and are very attractive for their easy implementation and optimisation, but they do not allow higher order dependence such as extremal tail dependence features. One way to generalise Gaussian process models so that higher order dependence can be handled, is to apply a transformation to the joint collection of Gaussian processes for each marginal time series model. We apply mean-variance transformation that results in the transformed variables having multivariate skew-t distributions and being finite dimensional realisations of a general multivariate skew-t process. Motivation for the Model Choice. There are numerous advantages of using Gaussian Processes, beginning with: ease of optimisation and interpretability of hyperparameters, flexibility, richness of covariance functions, allowing for various model structures. Using a likelihood ratio type test with a GP is a very natural choice, as estimating GP model parameters is often done on the basis of maximising likelihood, and therefore this estimation can be incorporated into the compound version of the likelihood ratio test (Generalised Likelihood Ratio Test, GLRT). From Gaussian variables, GPs inherited the property of being fully specified by the mean and the covariance, and so testing for model equivalence inherently means testing for equivalence of the mean and covariance functions. But many popular kernels do not have the ARD property, and using them for a likelihood ratio test settings gives no easy way to account for causal structures in covariance. Consequently, it is using GLRT with an ARD-GP that gives a uniformly most powerful test with an unparalleled flexibility: known asymptotic distribution under the null, explicit evaluation and in a closed form, and usefulness also for misspecified models. The proposed use of copula warping allows introduction of additional dependence, in particular tail dependence, while keeping the likelihood in closed form. Application. We provide a generic framework which can be applied to a wide range of problems, and which can be readily tailored or further extended. The illustrative examples included demonstrate how a range of data properties can be encoded in the model, and how they might affect the detection of causality. We present two real data application: to commodity futures data and inflation and interest rates. We show how the framework can be used in practice, and how it can be combined with, or enhance, more common approaches to analysing financial time series. Our observations are in line with financial interpretations, but they also offer additional insight and pose thought-provoking questions. Structure of the thesis. This thesis presents the research as it evolved: starting from an overview of a range of the causality methods already known, and demonstrating out why they are unsatisfactory. Subsequently, a new approach is presented -- a method based on Gaussian processes, that was developed to solve the drawbacks of the methods presented in the first part. Afterwards, an extension is proposed to widen the range of dependence structures, as well as marginal properties of the data that can be incorporated. Chapter 1 introduces the topic of the thesis, and reviews relevant literature. Chapter 2 discusses philosophical roots of the concept of statistical causality, as well as alternative notions of causality. After illustrating some of the varied ways of conceptual representation of causality, we present four distinct ways of modelling statistical causality. Chapter 3 contains background on the models considered: Gaussian processes, copulas and selected distributions. Chapter 4 describes inference procedures used: assessing hypothesis tests, generalised likelihood ratio test, permutation tests, and likelihood ratio test. The second part, New Perspectives on Causality Representation and Inference, presents the main contribution of our work. It starts with Chapter 5 containing the theoretical background for describing and testing causality with GP models. Chapter 6 extends the model from the previous chapter by introducing mean-variance transformation that results in a warped GP model, which can describe causality in the presence of skewness and tail dependence. Chapter 7 describes how synthetic data has been simulated, details the algorithm for approximating likelihood in the warped GP, and provides information on other relevant algorithms and the software used to implement our method. Chapter 8 presents an extensive experiment section, which aim to show, firstly, the good behaviour of the proposed procedures (model sensitivity and misspecification analysis), secondly, good power of the test for a range of structures, and, thirdly, the interaction of causality and tail dependence. Applications to real-world data are described in Chapter 9, where time series for commodities and currency markets are analysed. Finally, Chapter 10 presents the conclusions and directions for further development, and Appendices provide supplementary material.

In most inherited retinal degenerations such as age-related macular degeneration, death of photoreceptors leads to irreversible vision impairment. As the retina starts to degenerate, the retinal environment begins to remodel. While variations in the timing and magnitude of retinal response have been reported by various studies looking at different initiating insults, there has been no comparative assessment of the diseased retinae across time in multiple models of retinal degeneration. Here, the process of gliosis, outer nuclear layer cytoarchitecture and integrity of the outer limiting membrane (OLM) were examined in seven different mouse models of inherited retinal degeneration. Findings from this study demonstrated that each of the examined mutants manifests a unique pattern of alterations occurring in the retinal microenvironment, including significant changes to OLM junctions. Moreover, these animals showed that the type of the Müller glial response and its magnitude are not correlated with disease severity. Having identified marked but varied changes in the expression of GFAP, Vimentin and chondroitin sulphate proteoglycas (CSPGs) between different models, further work investigated whether it is possible to prevent or manipulate these changes. By knocking down expression of GFAP and Vimentin by with RNA interference (RNAi), the role of these proteins both in the development of degeneration-associated glial hypertrophy and in influencing photoreceptor transplantation outcome was assessed. Consistent with the idea that CSPGs deposition in the subretinal space represents a barrier to therapeutic interventions, an enzymatic approach was used to effectively digest these molecules in a diseased mouse retina. In the last instance, RNAi-mediated knockdown of CRB1, one of the key regulators of AJs, was used to introduce disruption of OLM integrity. The overall aim of this thesis was to provide a better understanding of the barriers in a diseased retinal environment and how they can be manipulated to ease in a future some of the therapeutic strategies to help to restore lost vision.

Auditory comprehension, the ability to understand spoken language, consists of a number of different auditory processing skills. In the five studies presented in this thesis I investigated both intact and impaired auditory comprehension at different levels: voice versus phoneme perception, as well as single word auditory comprehension in terms of phonemic and semantic content. In the first study, using sounds from different continua of ‘male’-/pæ/ to ‘female’-/tæ/ and ‘male’-/tæ/ to ‘female’-/pæ/, healthy participants (n=18) showed that phonemes are categorised faster than voice, in contradistinction with the common hypothesis that voice information is stripped away (or normalised) to access phonemic content. Furthermore, reverse correlation analysis suggests that gender and phoneme are processed on the basis of different perceptual representations. A follow-up study (same paradigm) in stroke patients (n=25, right or left hemispheric brain lesions, both with and without aphasia) showed that lesions of the right frontal cortex (likely ventral inferior frontal gyrus) leads to systematic voice perception deficits while left hemispheric lesions can elicit both voice and phoneme deficits. Together these results show that phoneme processing is lateralized while voice information processing requires both hemispheres. Furthermore, this suggests that commencing Speech and Language Therapy at a low level of acoustic processing/voice perception may be an appropriate method in the treatment of phoneme perception impairments. A longitudinal case study (CF) of crossed aphasia (rare acquired communication impairment secondary to lesion ipsilateral to the dominant hand) is then presented alongside a mini-review of the literature. Extensive clinical investigation showed that CF presented with word-finding difficulties related to impaired auditory phonological analysis, while functional Magnetic Resonance Imaging (fMRI) analyses showed right hemispheric lateralization of language functions (reading, repetition and verb generation). These results, together with the co-morbidity analysis from the mini-review, suggest that crossed aphasia can be explained by developmental disorders which cause partial right lateralization shift of language processes. Interestingly, in CF this process did not affect voice lateralization and information processing, suggesting partial segregation of voice and speech processing. In the last two studies, auditory comprehension was examined at the single word level using a word-picture matching task with congruent (correct target) and incongruent (semantic, phonological and unrelated foils) conditions. fMRI in healthy participants (n=16) revealed a key role of the pars triangularis (phonological processing), the left angular gyrus (semantic incongruency) and the left precuneus (semantic relatedness) in this task – regions typically associated via the arcuate fasciculus and often impaired in aphasia. Further investigation of stroke patients on the same task (n=15) suggested that the connections between the angular gyrus and the pars triangularis serve a fundamental role in semantic processing. The quality of a published word-picture matching task was also investigated, with results questioning the clinical relevance of this task as an assessment tool. Finally, a pilot study looking at the effect of a computer-assisted auditory comprehension therapy (React2©) in 6 stroke patients (vs. 6 healthy controls and 6 stroke patients without therapy) is presented. Results show that the more therapy patients carry out the more improvement is seen in the semantic processing of single nouns. However, these results need to be reproduced on a larger scale in order to generalise any outcomes. Overall, the findings from these studies present new insight into, as well as extending on, current cognitive and neuroanatomical models of voice perception, speech perception and single word auditory comprehension. A combinatorial approach to cognitive and neuroanatomical models is proposed in order to further research, and thus improve clinical care, into impaired auditory comprehension.

Anti-citrullinated protein antibodies (ACPA) are the main autoantibodies in rheumatoid arthritis (RA). Citrullination is the conversion of arginine to citrulline by peptidylarginine deiminase (PAD) enzymes. Periodontitis (PD) is a risk factor for RA, and there is evidence this may be due to citrullination by the unique prokaryotic PAD enzyme expressed by PD pathogen Porphyromonas gingivalis (PPAD). The aim of this work was to characterise PPAD, and investigate its potential role in the aetiology of RA. The 3D crystal structure of PPAD was solved, and functional residues identified. The catalytic domain comprises five ββαβ motif repeats, which create a canonical binding groove into the active site, and the immunoglobulin-like domain is an anti-parallel β-sandwich. PPAD lacks significant homology to human PADs, and forms distinct citrullinated peptides from those formed by human PAD isoforms PAD2 and PAD4 due to unique C-terminal substrate specificity. In vivo, immunisation with P. gingivalis proteins: PPAD, arginine gingipain (Rgp), and enolase (PgEnolase), induced ACPA responses in mice. In human cohorts, antibodies to PPAD, Rgp, and PgEnolase were increased in RA patients compared to osteoarthritis or healthy controls, and in RA patients with PD compared to those without. Six months of PD treatment saw a significant reduction in antibodies to P. gingivalis in RA patients with PD, but no change in ACPA. These results demonstrate P. gingivalis induces ACPA responses comparable to those observed in RA, potentially through formation of novel citrullinated peptides by PPAD. Structural and mechanistic information obtained here can be used in future studies targeting PPAD as a potential therapeutic strategy in RA.

Rh metal was implemented as a core component in the so-called three-way automotive exhaust catalyst (TWC) due to its excellent properties such as thermal stability, poison resistance and inert behaviour to react with any support materials. These catalysts are often loaded with small amount of promoter elements such as Ce and Zr that enhance their overall catalytic performances and catalyst lifetime. It is therefore desirable to develop new CeOx-based catalyst with both high redox activity and high thermal resistance. Samples of two surface coverages of rhodium on a non-porous high area alumina been synthesised; at 1.6 Rh and 4 wt % Rh loading. Ceria and zirconia have been deposited on the pre-supported Rh catalysts by the controlled surface reaction between the reduced Rh surface and an organometallic precursors: Zr(acac)4 and Ce(acac)3, with different cerium : zirconium ratios: Ce:Zr;1:0, 1:1, 2:1, 0:1. Ceriated Rh catalysts have been also prepared by a different synthesis method, wherein the ceria was deposited on γ-Al2O3 and subsequently Rh metal was added. A series of Rh catalysts produced has been characterised ex situ by BET surface area measurement, TEM, STEM HAADF, EDX, XPS, and XAFS. The combination of Diffuse Reflectance Infrared Fourier Transform Spectroscopy (DRIFTS), Mass Spectrometry (MS) and Energy Dispersive X-ray Absorption Fine Structure (EDE) techniques in a synchronous, time-resolved manner has been applied to determine the complementary information about structural and kinetic changes of the chemical system throughout CO exposure and two catalytic processes: CO oxidation and the reduction of NO by CO in the temperature range between 298 K to 573 K. It has been shown that advanced characterisation techniques, used in the time-resolved, in situ manner were needed in order to understand the behaviour of these complex catalytic systems under operating conditions. The effects of the catalysts’ composition, temperature and pulsed gas flow on catalyst performance have been effectively mapped by the combined DRIFTS/MS/EDE studies. The synthesis procedure has been found to be crucial to achieve the desired interaction between CeOx and Rh particles. No enhancement of ceriated Rh catalysts was observed when ceria was deposited directly on γ-Al2O3 (method I). The addition of each promoter element by the controlled surface modification procedure (method II) exhibits multiple effects on the catalysts performance. The DRIFTS/XAFS studies have confirmed that CeOx and ZrO2 facilitate the protection of Rh particles against extensive oxidation in atmospheres of air, O2 and CO. Larger Rh core particles of ceriated and zirconiated Rh catalysts have been observed when compared with undoped Rh samples. DRIFTS results indicate that by interaction of CO with Rh particles a significantly larger amount of linear CO species was formed on the ceriated Rh surface than in the non-ceriated Rh catalysts case suggesting the presence of predominant metallic Rh phase for that system. Moreover, only ceria doping was found to improve the catalytic activity of promoted Rh catalysts throughout CO oxidation in the temperature ramp mode under different (O2/CO) gas composition.

The collection, storage and distribution of water, managed by means of dams, reservoirs, tanks and wells, are activities central to life and religious ritual in South Asia, and occasion some of the subcontinent's most spectacular architectural conceptions and engineering achievements. This study is the first to address the subject of water architecture as a whole, to relate the structures of the various regions, contexts and types to each other, and to present a comprehensive interpretation of the history and meaning of South Asian water architecture. It draws attention to the architectural splendour and sacred associations of monuments, many of which have not been documented before, or which have been considered merely as technical constructions. As such, it is the first study to attribute to water architecture a central position within the corpus of South Asian architecture alongside and on equal rank with temple and residential architecture. The dissertation is a study of architectural structures relating to water in India, Nepal and Sri Lanka, mainly between the ninth and the nineteenth centuries. The structures under examination are divided into five main types: ghats (steps into water), tanks, kundas (deep stepped basins), wells and ornamental pools in palaces and water gardens. The dissertation shows how water structures signify both practical and metaphysical importance; it investigates the various forms and parts of water monuments, and it traces their development from simple to more complex forms of architecture. In particular, it is concerned with the shapes of the structures, which favour both secular and religious activities, express sacred and royal meanings, and provide a setting for the re-enactment of mythical events. The brief general introduction summarises the present state of research, discusses the sources and explains the chosen approach to the material. This is followed by an introduction to the religious meanings and cultural associations connected with water in the main religious traditions of South Asia. The five following chapters each deal with one of the five types of water architecture, and contain the main findings of the author's field-work. It is argued that the architectural framework of each of the principal types of water architecture is common to the entire subcontinent, that regionalism has considerably less influence on them than has hitherto been assumed, and that no type is exclusive to any one context. Each chapter analyses the main characteristics and the constituent architectural parts of the type, its variations, the border cases, and developments. The final chapter summarises the main results, examines common themes in water architecture, and outlines modem continuity in South Asia.

DNA replication is a fundamental cellular process. However, the structure and dynamics of the eukaryotic DNA replication machinery remain poorly understood. A soluble extract system prepared from Xenopus eggs recapitulates eukaryotic DNA replication outside of a cell on a variety of DNA templates. This system has been used to reveal many aspects of DNA replication using a variety of ensemble biochemical techniques. Single-molecule fluorescence imaging is a powerful tool to dissect biochemical mechanisms. By immobilizing or confining a substrate, its interaction with individual, soluble, fluorescently-labeled reactants can be imaged over time and without the need for synchrony. These molecular movies reveal binding parameters of the reactant and any population heterogeneity. Moreover, if the experiments are imaged in wide-field format, the location or motion of the labeled species along the substrate can be followed with nanometer accuracy. This dissertation describes the use and development of novel single-molecule fluorescence imaging techniques to study eukaryotic DNA replication. A biophysical characterization of a replication fork protein, PCNA, revealed both helical and non-helical sliding modes along DNA. Previous experiments demonstrate that the egg extracts efficiently replicate surface-immobilized linear DNA. This finding suggested replication of DNA could be followed as motion of the replication fork along the extended DNA. However, individual proteins bound at the replication fork could not be visualized in the wide-field due to the background from the high concentration of the fluorescent protein needed to compete with the extract’s endogenous protein. To overcome this concentration barrier, I have developed a wide-field technique that enables sensitive detection of single molecules at micromolar concentrations of the labeled protein of interest. The acronym for this method, PhADE, denotes three essential steps: (1) Localized PhotoActivation of fluorescence at the immobilized substrate, (2) Diffusion of unbound fluorescent molecules to reduce the background and (3) Excitation and imaging of the substrate-bound molecules. PhADE imaging of flap endonuclease I (Fen1) during replication revealed the time-evolved pattern of replication initiation, elongation and termination and the kinetics of Fen1 exchange during Okazaki fragment maturation. In the future, PhADE will enable the elucidation of the dynamic events at the eukaryotic DNA replication fork. PhADE will also be broadly applicable to the investigation of other complex biochemical process and low affinity interactions. It will be especially useful to those researchers wishing to correlate motion with binding events.

Aim: We sought to improve retention of neonatal resuscitation skills by modifying step 3 through additional functional verbalization in Peyton’s four step approach (P4S). Methods: Newborn life support (NLS) training was performed in a simulation-based setting. In contrast to the traditional approach, students taught with the modified approach were requested to explain every step of their performance in Peyton’s step 3. 123 Students were allocated into both experimental groups. Students were then assessed by megacode on day four (initial assessment) and 6 months (follow-up assessment). Results: Both groups showed similar scorings in the initial, follow-up assessment and in mean change. On initial megacode, time to start with initial inflation and post-resuscitation care was significantly faster in the control group. All showed a significant loss of performance irrespective of modification in step-3 in the follow-up assessment. Only time until start with post-resuscitation care shows a significant group difference in mean change between initial and follow-up with increasing time in the control and decreasing time span in intervention group. Conclusion: Both methods showed equal levels of knowledge acquisition and long-term decline in NLS performances. Verbalization in step 3 influenced speed of applied NLS performance.

This study examines the relationship between undergraduate music theory curricula and the widely varied backgrounds in written theory and aural skills brought to post-secondary institutions by entering students. A survey of Midwestern, large, public universities that offer a bachelor’s degree in music forms the basis for my discussion of how the curricula meet students’ needs. This analysis gives rise to specific proposals for music theory programs and curricula at these schools which, if implemented, can increase success rates of students. Advisor: Stanley V. Kleppinger