Search

This thesis investigates the manner in which the action sequences of contemporary Hollywood cinema reflect and constitute ways of imagining space. The thesis proposes that these sequences are highly spatialised presentations of bodily interaction with the world, and as such manifest cultural anxieties regarding the relationship between the individual and the built environment, and work to assure their viewers of the capacity of the human form to survive the disorienting spaces of contemporary architecture, globalisation and technology. In order to demonstrate this, the aesthetic and formal properties of action sequences are read alongside critical work exploring how space shapes social life, including influential texts by Henri Lefebvre, Michel de Certeau, Fredric Jameson and others. These readings reveal that both action sequences and critical spatial theory are similarly attentive to the difficulties, contradictions and possibilities of built space. A range of action sequences from Hollywood films of the last fifteen years, including sequences from Mission: Impossible - Ghost Protocol, The International, The Bourne Identity, The Bourne Supremacy, The Bourne Ultimatum, Jumper, Casino Royale, Quantum of Solace, Sucker Punch, Inception, Swordfish, The Matrix, The Matrix Reloaded, TRON: Legacy, Resident Evil, Resident Evil: Afterlife and Dredd 3D are analysed for how they depict space and spatial agency. Rather than concentrating upon the narratives of these films, the chapters of the thesis deal in turn with the ways in which action sequences express contemporary developments within the built environment; the consequences of globalisation; the impact of these spatial changes upon mental life; the challenges to bodily engagement raised by digital technology and cyberspace; and the modifications to representing space on film prompted by stereoscopic exhibition. Examinations of these sequences are used to build a model of the action sequence that suggests spatial appropriation and ideas around place-creation are crucial elements of the form.

Pregnant women undergo a series of metabolic and immunologic changes to ensure provision of nutrients to, and prevent rejection of, the fetus. To ensure continuous supply of glucose to the fetus, the mother increases glucose production, glucose intolerance and insulin resistance. To meet her own energy demands, the mother transitions from lipid storage to lipolysis. To prevent rejection of the fetal semi-allograft, the mother's immune system must be regulated, whilst maintaining protection against pathogens. Hypothesis: Well-recognised metabolic changes in pregnancy could impact maternal immune function. The aims of this project are to landscape the lipidomic profile using novel mass spectrometry techniques, and to determine whether monocytes undergo metabolic adaptation, if this occurs at 28 weeks of gestation, and if maternal obesity sabotages immunological adaptations. In addition, aims included investigation into the mechanisms which may protect the mother and fetus against SARS-CoV-2. Key findings unveiled significant phenotypic adaptations in the monocyte subsets during pregnancy, which are sabotaged by obesity. As the effect of maternal obesity is poorly understood, other immunological adaptations were investigated which revealed a shift to a Th1 and Th17 response which might contribute to the detrimental effects of obesity on pregnancy. At term, the monocytes illustrate a strong metabolic adaptation where their oxidative phosphorylation capabilities are reduced, confirmed by alterations in their mitochondria, with a downstream effect on their functionality with reduced production of lipid mediators and cytokines. While risk of infection with SARS-CoV-2 is low to pregnant women and the fetus, there is increased risk of preterm birth and admission into ICU. The fetus is relatively protected against infection, with cases of vertical transmission being rare. This thesis illustrates an elevated presence of soluble SARS-CoV-2 related molecules in breast milk and amniotic fluid which are postulated to act as decoy traps for the virus, which protects the neonate. In conclusion, this thesis has revealed novel findings into the immunometabolism adaptation to pregnancy.

Environmental targets require lower emissions, which necessitate increased operating efficiency for future generations of aeroengines. Consequently, higher operational temperatures will be needed, beyond the capabilities of current nickel-based superalloys. Therefore, new high temperature alloys are being investigated, including refractory metal high entropy superalloys (RSAs), based on the AlMoNbTaTiVZr system. These alloys comprise nanoscale order-disordered B2+bcc microstructures, similar to nickel-based superalloys, and have promising high temperature compressive yield strengths and competitive densities. However, RSAs typically have a ordered B2 matrix and can form Al-Zr-rich intermetallic phases, limiting room temperature ductility. This thesis aimed to develop understanding of the contributions of different elements to RSA microstructures, through the systematic study of constituent systems, which will aid the design of future RSAs. The nanoscale microstructures in RSAs are believed to form due to the miscibility gaps between the refractory metals and Zr. To investigate the contributions of different bcc+bcc miscibility gaps to RSA microstructures key simplified systems were studied. Nanoscale morphologies, like those in RSAs, are shown to form via a spinodal decomposition in the TaTiZr system, Chapter 4, albeit comprising of disordered bcc phases, but similar morphologies were not observed within the NbTiZr system, Chapter 5. Furthermore, compositional modifications in the TaTiZr system produced a refractory metal rich bcc matrix phase, indicating a potential route to produce more ductile RSAs. Through varying the ratio of refractory components in the NbTaTiZr and MoTaTiZr systems, Chapter 6, Nb was observed to lower the bcc+bcc solvus temperatures while Mo raised the solvus temperatures. The primary role of Al in RSAs was believed to be the ordering of the B2 phase. To investigate this premise, in Chapter 7, Al was added in a TaTiZr alloy with a Ti-Zr-rich matrix, analogous to complex RSAs, which demonstrated sufficient Al content can induce B2 ordering. However, B2 precipitates were only observed at relatively low temperatures of ∼700˚C, raising concerns for high temperature mechanical properties. The effect of Al was seen to be more complex in Chapter 8, where additions of Al into a TaTiZr alloy with a Ta-rich matrix increased the propensity for forming the nanoscale basketweave structure. In Chapters 8 and 9, where Al was removed from AlMoNbTaTiZr alloys, Al was observed to impact the volume fractions of the bcc phases formed. Critically, Al was also associated with the formation of intermetallic phases, the most prevalent of which is an Al-Zr-rich intermetallic related to the binary Al4Zr5 phase, which are believed to be deleterious to the mechanical properties. In Chapter 10, Mo suppressed some of the intermetallic phases in complex RSAs but a greater fraction of the Al-Zr-rich intermetallic formed. Microstructural stability of RSAs is critical to retain advantageous properties during high temperature service. Throughout this work, homogenised alloys were exposed to long duration thermal exposure at sub-solvus temperatures. In both simplified (Chapters 4-9) and complex RSAs (Chapter 10), the homogenised microstructures were not thermally stable but exhibited significant precipitate coarsening and many alloys formed additional phases. These studies highlight some of the challenges faced by RSAs and the potential for microstructural optimisation.

Ti-Nb based alloys show superelastic and shape memory properties over a wide range of temperatures, making them attractive for industrial applications. However, within the literature there are significant variations in the reported martensite start temperatures that have not been rationalised. To investigate the source of these discrepancies a model alloy, Ti-24Nb (at.%), has been studied in two microstructural conditions using a variety of analytical techniques. The evolution of the phase constitution of the material during a five step thermomechanical cycle was determined, in situ, using synchrotron radiation. To expand this dataset, the load-unload cycle was repeated at different temperatures in the range of -196°C to 30°C. The α'' martensite was observed to form when cold rolled material was cooled. In contrast, no evidence of the β to α'' transformation was observed in solution treated material, despite cooling to -196°C under identical conditions. Further transformation could be induced in both samples through the application of an external stress. These results highlight that the transformation behaviour in Ti-24Nb (at.%) is significantly influenced by microstructural condition. In light of these results, it was proposed that the martensitic transformation can be explained in terms of type I, II and III stresses rather than being solely reliant on the key transformation temperatures. To elucidate this, preliminary experiments investiging the effect of grain size and dislocation density on the transformation behaviour in Ti-24Nb-4Zr-8Sn (wt.%) were conducted. Changes in the grain size did not alter the transformation behaviour as significantly as the dislocation density, however, further investigations are required to validate this. Four Ti-Nb alloys were subjected to the same thermomechanical cycles, in situ, and the true martensitic start temperatures were determined and compared with previously reported datasets in the literature. The new total stress based mechanism was then used to rationalise why there is a significant amount of variation in the reported martensite start temperatures. Ti-Nb based alloys exhibit superelastic and shape memory properties over a wide range of temperatures, making them attractive for industrial applications. However, within the literature there are significant variations in the reported martensite start temperatures and associated behaviour that have not been fully explained. These variations are problematic to engineering design, thereby limiting the industrial uptake of these alloys. To investigate the source of these discrepancies a model alloy, Ti-24Nb (at.%), has been studied in two different microstructural conditions, cold rolled and solution heat treated, using a variety of different analytical techniques. Initial characterisation of the material was performed using X-ray diffraction and electron microscopy. Information relating to the evolution of the material in response to changes in stress and temperature were obtained in situ using synchrotron radiation. These diffraction data were used to directly characterise the phase constitution of the alloy during a five step thermomechanical cycle. To expand this dataset, this cycle was repeated a number of times to enable loading at different temperatures in the range of -196°C to 30°C. The α'' martensite was observed to form when cold rolled material was cooled. In contrast, no evidence of the β to α'' transformation was observed in material that had been solution heat treated, despite cooling to -196°C under identical conditions. Further transformation could be induced in both samples through the application of an external stress. In light of these results, it was proposed that the martensitic transformation that gives rise to superelastic behaviour can be explained in terms of type I, II and III stresses rather than being solely dependent on the key transformation temperatures. The observed results highlight that the transformation behaviour in Ti-24Nb (at.%) is significantly influenced by microstructural condition. To elucidate this further, the effect of grain size and dislocation density on the transformation behaviour in Ti-24Nb-4Zr-8Sn (wt.%) have been briefly investigated. The results indicated that the grain size did not alter the transformation behaviour as significantly as the dislocation density, however, these experiments were preliminary and further investigations into the microstructural features that change the superelastic properties exhibited by these alloys are required. The use of in situ synchrotron diffraction experiments are recommended to identify the true martensitic start temperature for an alloy in a given microstructural condition. Finally, the effect of niobium content on the true martensitic start temperatures and form of the stress-strain behaviour were obtained and compared with previously reported datasets in literature. The new total stress based mechanism was then used to rationalise why there is a significant amount of variation in the previously reported martensite start temperatures.

Natural selection favours organisms that are flexible in the face of changing environmental conditions, a trait that animals achieve by modifying their behaviour. Some of this flexibility arises from the activity of feeding related genes capable of controlling multiple behaviours in a functionally coherent way in response to environmental challenges. As such, these genes are promising targets for studies of the evolution of behaviour. However, while classical genetics can identify the genes themselves, it cannot find the alleles relevant for evolution in the wild. To study the natural variation of feeding behaviour, I investigated the satiety quiescence behaviour of Caenorhabditis elegans by recording a dataset of almost 5000 high-resolution videos of individual worms. The satiety quiescence behaviour is known to have significant genetic overlap with the mammalian postprandial somnolence behaviour, or 'food coma', so some of the genes are expected to be conserved and potentially provide insight into the human behaviour, as well. The almost 200 strains tested are from the Caenorhabditis elegans Natural Diversity Resource (CeNDR), a collection of wild type isolates that includes the full genome sequence of each strain in addition to other information. As satiety quiescence is a complex phenotype, I created an analytical pipeline that uses high resolution tracking and multidimensional phenotyping to break the trait down into its parts. This process is successful in identifying several new hypotheses about the way the behaviour itself is organised and controlled. The phenotypic components are then individually associated with genetic variants using genome-wide association mapping, revealing potential new genomic locations related to satiety quiescence.

Hesitancy in vaccination programmes has increased over the past few decades and has led to vaccination delays and refusals resulting in epidemic outbreaks and curtailing progress towards disease elimination. Addressing vaccine hesi- tancy is a public health priority. In this thesis we assess the global state of vaccine coverage and vaccine hesitancy by performing a number of timely anal- yses, providing key public health messages. These analyses can be used to address vaccine coverage and inform effective immunisation programmes and policies. In Chapter 2, we assess national-level childhood immunisation programmes by forecasting each nation's vaccine coverage rates. By these means we are able to provide estimates of likely future coverage levels and derive a quantitative measure of the strength of a country's immunisation programme - a measure we term the vaccine performance index. In countries where vaccines are freely available, our index captures losses to coverage based on hesitancy issues (non- access barriers that centre on personal beliefs and attitudes such as confidence in vaccines). We perform the first global-scale quantitative analysis of time- varying socio-economic determinants of vaccine coverage. In Chapter 3, we focus on direct measures of attitudes to vaccines. Using the largest survey ever conducted on attitudes to vaccines (almost 70,000 individ- uals across 67 countries), we examine global variation in (and socio-economic determinants of) attitudes to vaccines. This is achieved through a four-question survey on individuals' perceptions of vaccine safety, importance, effectiveness, and religious compatibility. In Chapter 4, we use a latent variable method to project these four attitudinal dimensions onto a single dimension which we claim to be the first quantitative measure of individual's vaccine hesitancy. This measure is termed the vaccine hesitancy index and we infer hesitancy for each individual. Hesitancy is investigated globally and we perform a case study of hesitancy in Nigeria, where there have been notable recent concerns with vac- cinations. Identifying countries with sub-optimal immunisation rates, vaccine-hesitant groups, and understanding socio-economic correlates of coverage and hesitancy enables us to understand the complex coverage-hesitancy interplay and address gaps in coverage.

The thesis presents various perspectives on physical and biological computation. Our fundamental object of study in both these contexts is the notion of switching/erasing a bit. In a physical context, a bit is represented by a particle in a double well, whose dynamics is governed by the Langevin equation. We define the notions of reliability and erasing time-scales in addition to the work required to erase a bit for a given family of control protocols. We call bits “optimal” if they meet the required reliability and erasing time requirements with minimal work cost. We find that optimal bits always saturate the erasing time requirement, but may not saturate the reliability time requirement. This allows us to eliminate several regions of parameter space as sub-optimal. In a biological context, our bits are represented by substrates that are acted upon by catalytic enzymes. We define retroactivity as the back-signal propagated by the downstream system when connected to the upstream system. We analyse certain upstream systems that can help mitigate retroactivity. However, these systems require a substantial pool of resources and are therefore not optimal. As a consequence, we turn our attention to insulating networks called push-pull motifs. We find that high rates of energy consumption are not essential to alleviate retroactivity in push-pull motifs; all we need is to couple weakly to the upstream system. However, this approach is not resilient to cross-talk caused by leak reactions in the circuit. Next, we consider a single enzyme-substrate reaction and analyse its mechanism. Our system has two intermediate states (enzyme-substrate complexes). Our main question is “How should we choose binding energies of the intermediates to minimize sequestra- tion of substrates (retroactivity), whilst maintaining a minimum flux at steady-state?”. Choosing very low binding energies increases retroactivity since the system spends a considerable proportion of time in the intermediate states. Choosing binding energies that are very high reduces retroactivity, but hinders the progress of the reaction. As a result, we find that the the optimal binding energies are both moderate, and indeed tuned with each other. In particular, their difference is related to the free energy difference between the products and reactants.

Mitochondria form an essential component of nearly all eukaryotic cells, are implicated in numerous diseases and may play important roles in ageing. Mitochondrial populations are dynamic, controlled and heterogeneous, with different types -- both mutant and wildtype -- potentially coexisting in single cells. This thesis will study the dynamics of both mitochondria and their genetic material (mtDNA) to improve our understanding of the role of these dynamics in pathology and ageing. This study suggests, as well as critically evaluates, reasons for the existence of complex continuous mitochondrial networks using coarse-grained mathematical models, underlining a nonlinear relation between functionality and network structure. Understanding the link between morphology and function is important as disruption of the former is directly implicated in cellular dysfunction. We perform experiments in which we measure the influence of mitochondrial fusion and division events on integrated mitochondrial membrane potential, an indicator of functionality, and find evidence for its conservation. The cellular homeostatic control acting on a mitochondrial population is poorly understood; to address this, we study the influence of general feedback control strategies on mutant and wildtype mtDNA dynamics. We introduce a simple linear control mechanism that captures a wide variety of biologically observed dynamics, and study optimal parameterisations through the construction of an energy-based mitochondrial cost function. Not only cellular control, but also gene-therapeutic control of mtDNA is studied, allowing us to investigate optimal treatment strategies to reduce mutant loads. The cellular proportion of mutant mtDNA molecules, known as heteroplasmy, is crucial in mitochondrial disease and we study the influence of cellular mtDNA exchange on heteroplasmy dynamics and mutant expansion during ageing. We find that this exchange of genetic material can induce preferential mutant expansion during ageing (even in the face of selection against mutants) through a stochastically driven increase in cellular mean heteroplasmy levels.

Observing networks completely is not always possible or practical. However, attributes of nodes can help us predict how they interact with one another---even when only partial observations are available. Based on this premise, we discuss four core concepts in this thesis. First, we model individuals as inhabitants of a social space spanned by their demographic attributes who connect with one another based on an unknown connectivity kernel. We develop the statistical techniques to learn the kernel from ego networks and infer friendship preferences from eleven survey datasets from the United Kingdom and the United States, as well as data we collected in three online surveys. Second, we use the kernel to develop an intuitive segregation measure applicable to discrete attributes, such as gender or ethnicity, and continuous attributes, such as age or income. For certain kernels, the measure reduces to a metric that allows us to quantify distance in a multidimensional social space. Third, we show that linear dynamics on large, spatially-embedded networks with a known connectivity kernel can be approximated by convolutional partial differential equations. We illustrate the spatial mean-field theory with two synthetic examples, demonstrate that coupled oscillators on networks behave wave-like at large scales, and test the approximation on a location-based social network. Finally, we develop a model-based community detection algorithm for time series. The hierarchical model allows us to propagate uncertainties from the raw data to community labels. We use the model evidence to automatically determine the number of communities and show that we can recover communities from simulated data even when the number of observations is small. We apply our approach to daily returns of the constituents of the S\&P 100 index to identify communities of stocks with similar behaviour.

Mitochondria are crucial organelles for complex cellular life: their dysfunction causes multiple devastating heritable diseases, and is associated with healthy ageing. Cells possess a population of mitochondrial DNAs (mtDNA) which are continuously turned over. Consequently, mtDNA mutations may arise and expand throughout the cell. Here, we aim to use stochastic modelling to deepen our understanding of mitochondrial populations, using inference techniques to connect with data, to provide rational hypotheses for biomedical interventions. Mitochondria are not static organelles: they form dynamic networks through organellar fusion and fission. We provide the first analytically-derived bridge between mitochondrial genetics and network dynamics. We show that the rate of increase in cell-to-cell variability in mitochondrial mutant load, as well as the rate of de novo mutation generation, is modulated by the fraction of mitochondria which are unfused from the mitochondrial network. We suggest that cells respond to rising mutant load of a particular mtDNA mutant by initially reducing in volume to maintain the density of unmutated mtDNAs and reduce power demand. When a minimum cell volume is reached, we suggest that cells then switch to power supply production through the induction of bioenergetic pathways. A non-linear relationship between cell size and mitochondrial functionality has recently been experimentally observed. We suggest that a metabolic scaling argument, combined with a simple model of cell death, plausibly accounts for the observed non-linearity. Finally, we explore how replication errors in nuclear DNA during neurodevelopment may seed regions of pathologically mutated neurons in the adult brain. By modelling neurodevelopment as a deterministic branching process, we infer the mutation rate during neurodevelopment from a recent dataset. We extrapolate from our model to suggest that pathological islands consisting of approximately 10^5 neurons may be common.

The purpose of this thesis is to study link cobordisms. The main tool we use to do so is given by Juhász’s cobordism maps, although we sometimes make use of a recent construction by Zemke. Given a decorated link cobordism, that is, a link cobordism with additional structure, there is a map induced at the level of link Floer homology (HFL, also denoted HFK for knots). When X is a decorated concordance from K_0 to K_1, we prove that the cobordism map F preserves the natural bigrading of HFK, and that there is a morphism of spectral sequences from HFK(K_i) to HF(S^3) which agrees with F on the first page, and with the identity of HF(S^3) on the infinity page. We use this result to obstruct the existence of invertible concordances between given knots, and to define a non-vanishing element of HFK(K) associated to a slice disc for the knot K. We then give a full description of the maps induced by elementary decorated link cobordisms, which generate all decorated link cobordisms. In particular, we relate the map associated to a saddle cobordism to a skein exact triangle which generalises Ozsváth and Szabó’s oriented skein exact triangle in HFL to decorated links. Lastly, we use this description to prove that the TQFT defined by HFL on a category of unlinks and cobordisms between them agrees with Khovanov’s reduced TQFT.

Cell-to-cell variation in expression of pluripotency- and lineage-determining factors has been proposed to be integral to the process of cell fate commitment in pluripotent cells both in vitro and in vivo. Understanding the sources of this heterogeneity in pluripotent stem cells promises greater insight into the mechanisms underlying cell fate choice. I identify mitochondrial membrane potential as an axis of heterogeneity in mouse embryonic stem cell populations, and show that high mitochondrial membrane potential marks cells that are in a stable self-renewing state. Partial overlap with previously described metastable subpopulations is demonstrated through gene expression analysis. I present evidence that similarly to previous findings in HeLa, heterogeneity in mitochondrial membrane potential is associated with variation in global transcription rate in mESCs. The direct impact of global transcription rate on differentiation propensity is demonstrated through manipulation of RNA Pol II transcription elongation rate. Mitochondrial variability is therefore likely a functionally relevant source of extrinsic gene expression variability in mouse embryonic stem cells.

Transplantation is the treatment of choice for end stage organ failure however current immunosuppressive therapies whilst effective at preventing acute allograft rejection, fail to prevent late graft loss due to chronic rejection and are associated with an increased risk of infection and malignancy. Therefore there is a clear unmet clinical need for improved strategies to prevent allograft rejection. OX40 is a member of the TNFR superfamily that has potent costimulatory properties. Although the impact of blockade of the OX40-OX40L pathway has been well documented in models of autoimmune disease, its effect on the rejection of allografts is less well defined. Therefore the aim of this thesis was to determine the impact of OX40 blockade on conventional and regulatory T cell responses to allografts. We found that activation of CD4+ and CD8+ naïve and memory T cells resulted in the induction of OX40 expression and that blockade of OX40-OX40L interactions partially inhibited the response of alloreactive T cells in vitro and prevented skin allograft rejection but did not result in the induction of tolerance. OX40 blockade was found to have no effect on the activation and proliferation of T cells but rather effector T cells failed to accumulate and migrate to skin allografts. This was shown to be the result of an enhanced degree of cell death amongst proliferating effector cells. In addition, blockade of OX40-OX40L interactions at a time of exposure to alloantigen resulted in a pool of Treg with an enhanced ability to suppress T cell responses to alloantigen in vitro and in vivo. Counter-intuitively, OX40 blockade was found to increase the potency of alloreactive Treg by promoting survival following re-activation. Finally, although OX40 blockade impacted both conventional and regulatory T cell responses, anti-OX40 administration did not promote skin or heart allograft survival in immunocompetent recipients and failed to synergise with blockade of other costimulatory molecules to prevent allograft rejection. In conclusion, these data demonstrate that blockade of OX40-OX40L interactions can attenuate naïve and memory T cell responses to alloantigen whilst promoting the survival of alloreactive Treg. Therefore, we propose that anti-OX40 would be a worthwhile adjunct to pre-existing strategies to induce tolerance.

In this thesis we propose some new approaches to the study of complex networks, and apply them to multiple domains, focusing in particular on protein-protein interaction networks. We begin by examining the roles of individual proteins; specifically, the influential idea of 'date' and 'party' hubs. It was proposed that party hubs are local coordinators whereas date hubs are global connectors. We show that the observations underlying this proposal appear to have been largely illusory, and that topological properties of hubs do not in general correlate with interactor co-expression, thus undermining the primary basis for the categorisation. However, we find significant correlations between interaction centrality and the functional similarity of the interacting proteins, indicating that it might be useful to conceive of roles for protein-protein interactions, as opposed to individual proteins. The observation that examining just one or a few network properties can be misleading motivates us to attempt to develop a more holistic methodology for network investigation. A wide variety of diagnostics of network structure exist, but studies typically employ only small, largely arbitrarily selected subsets of these. Here we simultaneously investigate many networks using many diagnostics in a data-driven fashion, and demonstrate how this approach serves to organise both networks and diagnostics, as well as to relate network structure to functionally relevant characteristics in a variety of settings. These include finding fast estimators for the solution of hard graph problems, discovering evolutionarily significant aspects of metabolic networks, detecting structural constraints on particular network types, and constructing summary statistics for efficient model-fitting to networks. We use the last mentioned to suggest that duplication-divergence is a feasible mechanism for protein-protein interaction evolution, and that interactions may rewire faster in yeast than in larger genomes like human and fruit fly. Our results help to illuminate protein-protein interaction networks in multiple ways, as well as providing some insight into structure-function relationships in other types of networks. We believe the methodology outlined here can serve as a general-purpose, data-driven approach to aid in the understanding of networked systems.

In this thesis, a highly comparative framework for time-series analysis is developed. The approach draws on large, interdisciplinary collections of over 9000 time-series analysis methods, or operations, and over 30 000 time series, which we have assembled. Statistical learning methods were used to analyze structure in the set of operations applied to the time series, allowing us to relate different types of scientific methods to one another, and to investigate redundancy across them. An analogous process applied to the data allowed different types of time series to be linked based on their properties, and in particular to connect time series generated by theoretical models with those measured from relevant real-world systems. In the remainder of the thesis, methods for addressing specific problems in time-series analysis are presented that use our diverse collection of operations to represent time series in terms of their measured properties. The broad utility of this highly comparative approach is demonstrated using various case studies, including the discrimination of pathological heart beat series, classification of Parkinsonian phonemes, estimation of the scaling exponent of self-affine time series, prediction of cord pH from fetal heart rates recorded during labor, and the assignment of emotional content to speech recordings. Our methods are also applied to labeled datasets of short time-series patterns studied in temporal data mining, where our feature-based approach exhibits benefits over conventional time-domain classifiers. Lastly, a feature-based dimensionality reduction framework is developed that links dependencies measured between operations to the number of free parameters in a time-series model that could be used to generate a time-series dataset.

Knowledge of protein sequences has exploded, but knowledge of protein function is needed to make use of sequence information, and this lags behind. A protein's function must be understood in context and part of this is the network of interactions between proteins. What are the relationships between protein function and the structure of the interaction network? In the first part of my thesis, I investigate the functional relevance of clusters, or communities, of proteins in the yeast protein interaction network. Communities are candidates for biological modules. The work I present is the first to systematically investigate this structure at multiple scales in such networks. I develop novel tests to assess whether communities are functionally homogeneous, and demonstrate that almost every protein is found in a functionally homogeneous community at some scale. The evolution of protein sequences is well-studied, but comparatively little is known about the evolution of protein function. Such knowledge is needed to understand when it is appropriate to annotate newly sequenced proteins by transferring functional information from homologs-i.e. evolutionarily related proteins. In the second part of my thesis, I assess the success of transferring protein-protein interactions across species and use this to estimate the rate at which interactions are lost in evolution. At levels of sequence similarity associated with functional annotation transfer, I demonstrate that protein-protein interaction transfer is unreliable. The relevance of community structure for understanding protein function and the low conservation of individual interactions, suggests a possible role for communities in the evolution of cellular function. I discuss this possibility in my conclusions.

Ecosystems are often made up of interactions between large numbers of species. They are considered complex systems because the behaviour of the system as a whole is not always obvious from the properties of the individual parts. A complex system can be represented by a network: a set of interconnected objects. In the case of ecological networks and food webs, the objects are species and the connections are interactions between species. Many complex systems are dynamic and exhibit intricate time series. Time series analysis has been developed to understand a wide range of natural phenomena. This thesis deals with the structure, dynamics, and robustness of ecological networks, the spatial dynamics of fluctuations in a social system, and the analysis of cardiac time series. Biodiversity on Earth is decreasing largely due to human-induced causes. My work looks at the effect of anthropogenic change on ecological networks. In Chapter Two, I investigate predator adaptation on food-web robustness following species extinctions. I identify a new theoretical category of species that may buffer ecosystems against environmental change. In Chapter Three, I study changes in parasitoid-host (consumer-resource) interaction frequencies between complex and simple environments. I show that the feeding preferences of parasitoid species actively change in response to habitat modification. Ecological networks are embedded in spatially-heterogeneous landscapes. In Chapter Four, I assess the role of geography on population fluctuations in an analogous social system. I demonstrate that fluctuations in the number of venture capital firms registered in cities in the United States of America are consistent with spatial and temporal contagion. Understanding how physiological signals vary through time is of interest to medical practitioners. In Chapter Five, I present a technique for quickly quantifying disorder in high frequency event series. Applying the algorithm to patient cardiac time series provides a rapid way to detect the onset of heart arrhythmia. Increasingly, answers to scientific questions lie at the intersection of traditional disciplines. This thesis applies techniques developed in physics and mathematics to problems in ecology and medicine.