Search

Enveloped viruses hijack the protein production machinery of their host in order to proliferate and their replication/infectivity relies on the correct folding of their surface glycoproteins. This dependence on glycoprotein quality control for viral proliferation makes the host endoplasmic reticulum quality control (ERQC) mechanisms an attractive target for the development of host-targeting antiviral agents, which may show broad-spectrum activity and a high genetic barrier to resistance. A key ERQC protein is α-glucosidase II (GluII), a heterodimeric enzyme bearing a catalytic α-subunit of the GH31 family and an accessory β-subunit that is necessary for full catalytic activity and for the localisation of the heterodimer in the ER. GluII mediates quality control of the folding of nascent glycoproteins by trimming glucose residues from its substrate glycan Glc1-2Man9GlcNAc2. Inhibition of GluII can be achieved using iminosugars which mimic the glucose residue of the native substrate, causing the build-up of misfolded protein in the ER and prevents the secretion of correctly folded glycoproteins. In the case of enveloped virus infection, this inhibits virion secretion or reduces infectivity of secreted virions and has been shown to elicit antiviral effect against a range of enveloped viruses in vitro and in vivo. Iminosugar inhibitors of GluII are effective, but have been associated with significant drug-related adverse effects. In this thesis, I present a novel approach to the inhibition of GluII that involves disruption of the heterodimeric protein-protein interface within the enzyme. Removal of the accessory β subunit from the catalytic α-subunit has been shown to decrease catalytic activity and may induce the export of the catalytic domain out of the ER. Thiopyridones were discovered as small molecules that interfere with the α/β interface and I investigate this protein ligand interaction using techniques such as mass photometry, mass spectrometry and high performance liquid chromatography among others to confirm the capability of these molecules to disrupt the α/β interface and potentially inhibit GluII. Additionally, I outline the design and synthesis of novel iminosugar compounds with the aim of improving the selectivity for GluII relative to existing best-in-class iminosugars. I also use molecular dynamics for in silico examination of the full extent of the GluII α/β interface and the roles of peripheral domains of the β-subunit that are not fully understood based on existing structural data.

Complex observation processes abound in ecology and epidemiology. In order to answer the large-scale, urgent questions that are the focus of modern research, we must rely on indirect and opportunistic observation. Relating these data to the biological processes we are interested in is challenging. Statisticians working in this area need an understanding of both state-of-the-art modelling techniques and the field-specific nuances of how the data were generated. As a result, many methods to deal with complex observation processes are highly bespoke. Bespoke models are hard to translate between contexts and, because they are often presented in field-specific language, hard to learn from. Modelling of observation processes is thus a fractured area of study, leading to duplication of research effort and limiting the rate at which we can make progress. In this thesis, I aim to provide a road-map to how we might achieve some unification in this area. I begin by establishing a conceptual framework that can be used to describe observation processes and identify methods to address them. The framework defines all observation processes as a combination of issues of latency, identifiability, effort or scaling (L.I.E.S.). I illustrate the framework using motivating examples from ecology and epidemiology. The risk with conceptual frameworks is that they can be over-fitted to existing data and may fail when faced with new, real-world problems. To address this, I also approach the problem from a bottom-up perspective by tackling a series of ecological and epidemiological case studies. Each case study requires novel statistical methods to deal with the observation process. By developing new methods, I explore the world of observation processes potentially not well-captured in the literature. I then explore whether these case studies motivate revision or reassessment of my conceptual framework. While the case studies were chosen to challenge the L.I.E.S. framework, I find that they mutually reinforce each other. The framework provides a helpful scaffolding with which to describe the problems in the case studies. The case studies provide useful examples of more complex observation processes and how the four issues encoded in L.I.E.S. interact with one another. These findings illustrate the value of a framework for unifying approaches to observation processes.

This thesis describes an approach to the identification, design and synthesis of novel fluorine containing heterocyclic fragments to be used as starting points for FragmentBased Drug Discovery programmes. Heterocyclic motifs are common in medicinal chemistry due to their robust synthesis and available chemical transformations. Fluorine incorporation into heterocyclic structures can modulate many physicochemical properties of a compound, not least its ability to block a metabolically labile hotspot. (Chapter 1). Herein is detailed the filtering of a generated database (GDB) of 26 million compounds to identify a dataset of 11,596 unreported, undocumented fluorinated heterocycles. With the use of a novel elaboratability programme (developed by James Davidson at Vernalis Research Ltd.), we were able to rank the fragments in terms of how readily they could be elaborated. Additionally, with the use of Pipeline Pilot, we were able to calculate the physicochemical properties of the dataset and compare it to commercial fragment libraries which showed that the fragments followed the "Rule of 3" guidelines for fragments. Applying a principal moment of inertia (PMI) plot for the dataset and comparing it to commercial libraries, we demonstrated that we are exploring underreported chemical space. (Chapter 2). The synthesis of four of these identified fragments is detailed, including reaction optimisation using design of experiment (DoE) approaches as well as efforts for late-stage fluorine incorporation via both traditional and newer methods, such as C - H functionalisation. (Chapter 3). Lastly, is the development and optimisation of a new photo-redox assisted decarboxylative fluorination strategy of N-heterocyclic acetic acids to their corresponding fluoromethyl counterparts. (Chapter 4).

Electric propulsion has the potential to induce a paradigm shift in how aircraft perform, provided the same shift occurs in their design. Improvements to conventional aircraft have reduced CO2 emissions per passenger kilometre by, on average, 3.8% per year since 1960, but these incremental improvements to conventional designs are not sufficient to meet the ambitious goals for reductions in the environmental impact of air transport. To meet these goals, radically different architectures and propulsion systems are required. These radical architectures are by definition difficult to design due to the lack of historical data to enable configuration choices, normally enabled by prior knowledge of performance trends for various configurations. Without the configuration established, optimisers cannot be used due to the small integer variable choices, such as number and position of propulsors, embedded in configuration design. This challenge is solved by developing a optimisation scheme for hybrid-electric aircraft that can deal with the small value integer variables that derail conventional schemes. This scheme is applied to conventional aircraft, for conventional missions, and replicates the expected performance and design of these aircraft. Then, the scheme is applied to missions currently infeasible with conventional aircraft. First of these missions was a STOL aircraft with 500 km range carrying 10 people. Application of this optimisation scheme produced an aircraft capable of a 241 m take-off run, cruise range of 500 km at 200 kt, burning only 287 kg of fuel. Second of these missions, a STOL survey aircraft designed to operate in the High Himalayas, an aircraft that bridges the capability gap between the remote access of helicopters and the endurance of fixed wing aircraft, coupling a 250 m balanced field length with a 1200 km range whilst burning only 217 kg of Jet-A. This work shows how the unlocking of this integer variable problem is critical to evaluating the true potential of electric propulsion architectures and how they shift and morph conventional design space maps, and unlock performance previously unavailable to both conventional designs and conventionally configured aircraft with electric propulsion.

This thesis aims to understand how methodological and conceptual approaches to complexity in quantitative analysis can improve evidence and decision-making, specifically for schistosomiasis control in Uganda and more broadly within global health. Engaging directly with the complexity through methodological choices provided new insights into policies and practices in global health. In Paper 1, I provided an overview of actors and power dynamics in global health, by describing the changing landscape of global health actors as it relates to relative shifts in power over time. This is accomplished by capturing the emergent, dynamic network structure of development aid for health in the period encompassing the 'MDG era', between 1990 and 2015. This paper was published in the Journal of Health Policy and Planning (https://doi.org/10.1093/heapol/czac025). Paper 2 aimed to develop evidence for decision-making in response to the needs of policymakers and practitioners, with a focus on schistosomiasis transmission and control activities in Uganda. This was accomplished by (1) capturing the perspectives of national and sub-national decision-makers on schistosomiasis transmission using participatory modelling, and (2) using the participatory modelling outputs to inform mathematical model simulations in response to the evidence needs. The implementation of this approach challenged the balance of power between international and domestic actors in the development of evidence and decisions regarding the delivery of global health interventions. This paper was published in BMJ Global Health (http://dx.doi.org/10.1136/bmjgh-2021-007113). Paper 3 used the outcomes of the participatory systems mapping workshops and individual-based simulations to guide the scope and content of economic evaluations of schistosomiasis interventions. The results indicated that the most cost-effective scenario is a system of implementation reliant on volunteers from within communities and donated drugs. As anticipated, when all else is held equal, including these costs result in lower cost-effectiveness ratios relative to other interventions. Further, the results bring into question the purpose of continuing interventions which are not predicted to achieve the desired targets within the 30-year time horizon. This paper highlighted potential opportunities for schistosomiasis intervention design and implementation which is more aligned with the aims of equitable, country-led sustainable development. Paper 4 shifted the focus within the discussion of evidence for decision-making in global health to consider one particular type, peer-reviewed publications, which is most often considered as 'best practice' in evidence-based decision-making. A systematic review captured the network of authors who had published on MDA. These results constituted the sampling frame for a remote survey to elucidate perspectives on their roles in policy and practice related to MDA. The findings highlighted the ongoing structural disparities in research leadership and found broad concern about opportunities and about disconnects that limit engagement between researchers and decision-makers for use of primary research in policy and decision-making processes. Paper 4 was published in the Journal of Public Health Policy (https://doi.org/10.1057/s41271-021-00294-x). Broadly speaking, the papers in this thesis have shown that while reductionist, linear perspectives may be part of the reason for the continuation of ineffectual policies and practices, the confluence of politics, power relations, and economies in the context of a complex system of actors and processes also plays a significant role with regards to policy and practice decision making. This was observed in relation to schistosomiasis in Uganda and more broadly in global health at the system level. This thesis uses language and methods common in health sciences to communicate critiques in a way that can be engaged with by health policy-makers, practitioners, and many public health researchers. Finally, this thesis showed the possibilities for using network-based and computational models for understanding complexity within the global health 'system'.

Active matter systems are comprised of bodies that consume energy from the environment for self-propulsion and as a result are driven out-of-equilibrium. These systems encom- pass a range of lengthscales and describe both living and artificial systems. Moreover, the motion and collective interactions of these active bodies lead to an ensemble of rich and complex behaviours not seen in equilibrium systems. On mesoscopic lengthscales (nm to μm), the prominent active systems are bacterial swimmers and active colloidal particles. However, as a result of the experimental challenges involved, there have been relatively few studies of mesoscale active matter systems in 3D. In this thesis, we aim to address two poorly understood areas of active systems in 3D: the crystallisation of active Brownian spheres, and the behaviour of active matter in 3D non-trivial confinement. Regarding crystallisation, we perform Brownian dynamics simulations to explore nucleation and polymorph selection in crystals of mono-disperse active spheres. We find two activity de- pendant crystallisation regimes: spinodal, and nucleation and growth. In the former we observe an increase in the crystal growth rate due to the accelerated dynamics in the melt, and in the latter we find that activity inhibits nucleation. Remarkably, activity also has a strong effect on polymorph selection. While the passive system crystallises in an equimolar mixture of FCC and HCP, active particles progressively favour the FCC phase at higher activity. The transport of mesoscale active matter occurs naturally in complex environments that are often crowded, random and irregular. We explore structural and dynamical properties of 3D active particles in random environments with a two-pronged approach. First we conduct Brownian dynamics simulations of active Brownian spheres in two random environments. We find that the confinement increases the heterogeneity of the dynamics, with new populations of absorbed and localised particles appearing close to the obstacles. This heterogeneity has a profound impact on the motility induced phase separation exhibited by the particles at high activity, ranging from growth in random disorder to a bicontinuous phase separation in porous environments. Finally we approach this same problem through experiment and present the first 3D active colloidal system in non-trivial confinement. The system is comprised of active Janus particles driven via an external field, propelling through the porous network surrounding a colloidal gel. This system is imaged in real-time with confocal microscopy and the particle dynamics are extracted with particle tracking methods.

This thesis explores the ways in which Simondon's transductive ontology both engages with and is made possible by natural scientific thinking. It is argued that Simondon's conception of transduction signals the necessity for an engagement between ontology and contemporary science and makes a number of significant contributions to this task - in particular, regarding his conceptions of energy and relation. In part, we attempt to demonstrate and defend his philosophy of transduction as a fruitful engagement with natural science, whilst equally it is argued that a number of aspects of this relationship are problematic. The general and guiding problem of this thesis is the possibility of an ontology in light of contemporary scientific thinking and discovery. Simondon's conception of transductive individuation is a significant attempt in this regard, as it rethinks ontology and individuation in light of theories, concepts and descriptions derived from contemporary science. This thesis attempts to situate Simondon's ontology relative to direct influences on his work and texts from the scientific and philosophical context in which he was writing. Close attention is paid to his engagements with the science of energy, concepts of relation in twentieth-century biology, epistemology and ontology, and cybernetic articulations of homeostasis and information. In this way, we equally contend that scientific thinking is not of mere conceptual inspiration for transduction, but instead offers ontological descriptions on which Simondon's ontology rests. As the argument progresses, problematic or unthought aspects of the relationship between transduction and scientific thinking are emphasised. Discussing Simondon's thought in light of that of Bergson and Deleuze, we emphasise and problematise the differences between their expressions of singularity and their relationships to scientific thinking. The primary problem, in this regard, has to do with the priority and nature of scientific thinking: if thought comes first, constituting and confirming Simondon's transductive philosophy and Deleuze's ontology of difference, then it may also constrain individuation or creativity to scientific concepts. Equally, if scientific thought is developing through a process of error and rectification, then it may not be the stable foundation articulated and required by ontologies such as those of Simondon and Deleuze.

This thesis argues that Victorian poets were persistently drawn to the experience of partial, uncertain, or elusive knowledge, situations in which knowing is felt to be simultaneously tantalising and potentially entrapping or chimerical: a set of conditions I group under the term 'unknowing'. Exploring the ways in which the idea of knowledge in the nineteenth century was saturated in value and feeling, I show how the resistance of certain objects to being fully known (people, God, art, nature, the future) could generate not merely frustration and disappointment but fresh sources of possibility and pleasure. For the poets I study, such concerns were intimately linked to the question of poetry's own relationship to knowledge. In the fibres of their poetry, as well as in the writings of some of the period's most inventive thinkers, I trace a vital and availing tradition which stresses how poetry may sharpen our awareness of that which exceeds or escapes our knowing. My first chapter reads the poetry of Arthur Hugh Clough as a sustained dialogue with Ralph Waldo Emerson's critique of 'knowingness'. My second connects Robert Browning's interest in the natural world with notions of foiled or occluded knowledge, drawing on the writings of John Ruskin. My third chapter proposes two ways of thinking about knowing in the poetry of Dante Gabriel Rossetti: an aesthetics of vagueness stemming from Renaissance art theory, and the dreamlike phenomenology of déjà vu. My fourth chapter pursues Alice Meynell's interest in God's hiddenness as a potential resource of faith, linking it to the thought of John Henry Newman and the figure of the Virgin Mary. My final chapter explores Thomas Hardy's attention to the difficulties of knowing (and being known by) other people, showing how more fruitful modes of engagement may depend upon relinquishing the desire for secure knowledge.

In this thesis we investigate the structure of quantum many-body systems. We explore how quantum systems evolve on their own, how the classical environment can influence, impair, inhibit the dynamics of controlled quantum systems, how entanglement grows in closed quantum systems, and how tools from the study of classical chaos can help us to characterise and quantify this growth. We propose tools for controlling quantum systems to achieve useful tasks; we develop algorithms for the quantum simulation of other, infinitely large quantum systems, making use of a formalism - tensor networks -familiar from classical studies of quantum behaviour. We investigate how inspiration from physics can be used to develop further quantum and classical algorithms, in quantum simulation, machine learning, and beyond.

Deep learning methods have experienced a revolution, driven by their successful application in fields such as computer vision and natural language processing. In this thesis, we describe several novel methodologies leveraging deep learning for applications to pre-clinical drug discovery. First, we propose a generative approach to the design of molecular linkers which incorporates basic 3D information. In large-scale tests, we find that our method substantially outperforms a database-based approach, the previous de facto approach for this problem. Through a series of case studies, we demonstrate the application of our approach to scaffold hopping, fragment linking and PROTAC design. We then extend this framework to incorporate physically meaningful 3D structural information, providing a richer prior for the generative process, and also apply our method to molecular elaboration tasks, such as R-group design. We then turn our attention to predictive modelling, in particular structure-based virtual screening. We find that the advances in convolutional neural networks (CNNs) for general computer vision tasks are applicable to structure based virtual screening. In addition, we propose two techniques to incorporate domain-specific knowledge into this framework. First, we show that limitations in docking necessitate the use of multi-pose scoring and demonstrate the benefits of an average scoring policy. Second, we propose a transfer learning approach to construct protein family specific models, utilising knowledge of the differences between protein families. Finally, we investigate how a generative approach can be used to improve the training and benchmarks sets employed in structure-based virtual screening. We propose a deep learning method that generates decoys to a user's preferred specification in order to control decoy bias or construct sets with a defined bias. We show that our approach significantly reduces the bias contained in such sets. We validate that our generated molecules are more challenging for docking-based approaches to separate from bioactive compounds than previous decoys. In addition, we show that CNN-based structure-based virtual screening methods can be trained on such compounds.

Design, development and deployment of cyber-physical systems (CPS), in particular human-centric CPS, is growing at a rapid pace, with the continuing downward trend in silicon costs and increasing demand for automation in the home, office, cities and national infrastructure, to cut costs, increase efficiencies and use less natural resources. Human-centric CPS can play pivotal a part in improving human well-being and safety, improve security and emergency evacuation, as well as helping us meet the worldwide targets for climate change set out by the Paris Agreement, involving the development and deployment of smart-homes, -cities and -countries developed by government, industry and academia, to reduce energy and resource waste. However, simulating and analysing a human-centric CPS remains a challenging task, involving simulation of distributed systems in tandem with human-mobility, real-world physics and phenomena. Instead these simulations are often carried out in distinct disconnected tools, resulting in poor end-to-end testing and disjointed analysis of the CPS in different scenarios. Cyber-physical systems interact with the physical world, hence, poorly tested systems can cause more harm than good, damaging infrastructure, harming people and even risking lives when systems fail. Existing tools for simulating and analysing human-centric cyber-physical systems do not support real-time simulation of environments with dynamic human-mobility and phenomena-on-demand, instead only supporting static scenarios using recorded traces. In this thesis we present a novel cyber-physical system simulation framework for the simulation and analysis of human-centric cyber-physical systems. The framework utilises a video game engine to simulate the real world, create human-mobility of individuals and crowds that adapt to the environment, and generate phenomena-on-demand. The CPS is simulated using the Cooja WSN simulator, providing accurate and scalable simulation of deployable code. Using the virtual world within the game engine, we've developed a novel approach to analysing simulations, coined ''visual diffing'', enabling developers to visually compare the physical and virtual activity of two simulations simultaneously intuitively within the context of the visual simulation. We demonstrate the framework through two human-centric CPS case studies, which evaluate the problems of presence-based office lighting and dynamic fire evacuation navigation. Both problems rely upon the accurate simulation of the virtual environment, the human agents within it and the interaction between the cyber, physical and human aspects of the system. We demonstrate that the use of a game engine to simulate the real world, model human-mobility using behaviour trees and enable visual diffing, provides a feasible way to simulate and analyse human-centric CPS to test and discover differences between different simulated scenarios. We also evaluate the performance and trade-offs of realising this approach.

The following critical appraisal presents eight interlinked works that apply and extend Michael Barkun’s (2003) concept of ‘improvisational millennialism’. This body of work demonstrates that, as Barkun suggests, the concept is widely applicable to the online communities in which stigmatized knowledge is widely accepted. Moreover, it extends the definition to consider how improvisational millennialism provides ill-defined or dispossessed communities a means of articulating a collective relationship to historical time as well as a crude means of shoring up basic assumptions of group membership. Mythical pasts and millennial expectation provide the opportunity for shared eschatological orientation whilst the dualism of conspiracy theories demarcates between the communities and their outsiders. This critical review demonstrates how the journal articles and book chapters collected in the appendices provide specific examples of the application and extension of improvisational millennialism. The examples chosen are varied but a persistent theme drawn out through analysis is the role that national cultures – official and official – are articulated through improvisational millennialism. The examples include consideration of how the depiction of millennial beliefs in the mass media contribute to national cultural constructs but more typically focus on the use of improvisational millennialism in online communities. Of the latter, the greater number of examples are concerned with improvisational millennialism within the neo-fascist milieu. Mobilised by conspiracy theories with apocalyptic subtexts, the far right reliance on improvisational millennialism demonstrates the implicit danger of the increased incursion of stigmatized knowledge into the cultural mainstream. This critical review serves to show that despite being typified by a syncretic bricolage of unconnected ideas and traditions, improvisational millennialism is reflective of both social and political realities.

Structure-based drug discovery uses information about the structure of a protein to identify novel ligands that bind to the protein. The fundamental problem in structure-based drug discovery is predicting if, how, and how strongly a possible ligand binds to a protein. This is often accomplished using scoring functions to rapidly estimate the strength with which a ligand binds to a protein -- its binding affinity. This thesis explores the use of machine learning techniques to improve scoring functions for protein-ligand binding affinity. We first analysed the features used by several published machine learning scoring functions, before showing that augmenting these features with ligand-based features can improve scoring function performance. We then compare the performance of different machine learning algorithms. We next perform a series of experiments to investigate how the size and composition of the training set, and its similarity to the test set, influences the performance of Random Forest scoring functions. We find that regardless of training set composition, augmenting structure-based feature sets with additional ligand-based features leads to enhanced scoring function performance on a diverse test set. We further investigate the predictions of a Random Forest using only ligand-based features, and find that, when a ligand has different binding affinities for multiple binding partners, this ligand-only model is predictive of the mean binding affinity of a ligand for its binding partners. Finally, we address the use of docked poses for the ligand instead of experimentally-determined binding modes. We find that pose prediction errors are common. We show that using docked poses in place of crystallographic binding modes reduces scoring function performance, and that augmenting a structure-based scoring function with ligand-based features can help to counteract this effect. We then construct a new data set and show that generalising to new data and novel targets remains challenging for machine learning scoring functions. In this thesis we examine whether the use of a more detailed representation of the physicochemical properties of a ligand can improve machine learning scoring functions for protein-ligand binding affinity.

The majority of Earth's surface topography is controlled by variations in crustal and lithospheric thicknesses and densities that are principally generated at plate boundaries. Nevertheless, it is recognised that some fraction of topography is maintained by convective circulation of the Earth's interior. Geodynamic simulations typically predict that this dynamic component of surface topography varies on wavelengths of ~ 10,000 km with amplitudes of up to ~ ± 2 km. However, recent compilations of oceanic residual depth measurements reveal topographic anomalies on much shorter wavelengths of ~ 1,000 km. Limited observational constraints exist regarding the spatial and temporal evolution of this dynamic topography on the continents. In this dissertation, I explore two ways in which such constraints might be obtained from fluvial landscapes and from sedimentary basins. I first investigate the growth of the Anatolian Plateaux. Elevated Miocene marine sedimentary deposits distributed across Anatolia indicate that significant uplift has occurred during Neogene and Quaternary times. However, the precise timing and spatial pattern of uplift are not well constrained. Fluvial drainage networks appear to contain quantitative information about the spatial and temporal evolution of continental topography. Such information is extracted from Anatolian drainage networks using an inverse modelling approach. Separately, estimates of asthenospheric temperature were obtained by modelling geochemical compositions of mafic volcanic rocks. Results suggest that elevated topography is supported by anomalously hot asthenosphere and by a thinned lithospheric plate. Regional uplift and magmatism appear to have propagated from east to west since ~ 20 Ma. Consistent patterns linking uplift, magmatism and asthenospheric temperature imply that convective processes have had an important influence on the spatial and temporal evolution of the Anatolian Plateaux. Secondly, I focus on the Surprise Canyon Formation, a Carboniferous deposit that is exposed in the Grand Canyon of western North America. This formation, which contains terrestrial flora, consists of isolated channels incised into the underlying marine Redwall Limestone. These channels are overlain by the marine Supai Group. This stratigraphic relationship requires an episode of transient base-level fall. The chronology and evolution of the Surprise Canyon Formation is reviewed and revised, in order to compile a regional database of subsidence histories. Emplacement and subsequent cooling of warm asthenospheric material can explain significant stratigraphic relationships. Thirdly, I explore the structure and evolution of the intracratonic Congo sedimentary basin. This enigmatic basin's protracted stratigraphic record, which includes periods of rapid sedimentary acculation separated by long unconformities, remains poorly understood. Seismic reflection profiles acquired during the 1970s are re-analysed in order to improve the stratigraphic architecture. Subsidence histories for four wells are calculated and modelled. I show that these histories are consistent with basin initiation during a Neoproterozoic-Cambrian phase of rifiting followed by a long phase of thermal relaxation, a consequence of the basin's location on thick cratonic lithosphere. This post-rift phase is punctuated by periods of uplift and denudation that can be explained by emplacement and cooling of warm asthenospheric material at the base of the plate. I conclude that quantitative information about the dynamics of the Earth's mantle can be obtained by careful analysis of landscapes and the stratigraphic record.

To advance important social values it is often necessary to change the law. Yet changes in the law create not only 'winners', but also 'losers'. Paradigmatic examples are workers and corporations in the coal industry that are adversely affected by climate change laws. Should governments take steps to avoid or mitigate transitional losses and gains? If so, for whom and under what conditions, and why? This is the normative 'problem of legal transitions', the subject of this thesis. Various theories to resolve the problem have been proposed by scholars in the normative branches of political, legal and economic theory. The first part of this thesis critically evaluates the four dominant families of theories: property theories (in the classical liberal and libertarian tradition); efficiency theories (in the 'law and economics' tradition); legitimate expectations theories; and justice theories (the last two in the liberal-egalitarian tradition). The 'wide reflective equilibrium' method is employed to evaluate both the implications of each theory in concrete cases and the antecedent (e.g. ontological) commitments in which it is grounded. The existing theories-which tend to entail state responses that are either extremely conservative (e.g. full grandfathering or compensation) or purely reformative (let all losses and gains lie where they fall)-are found wanting. The second part of the thesis specifies and defends a new theory-Adaptive Responsibility Theory-which forges a principled middle path. The appropriate state response is shown to be a function of two kinds of reasons: wellbeing reasons and fairness reasons. Crucially, these reasons have both a direction (conservative vs reformative) and a magnitude/weight. This opens up the wide middle ground between the extremes, justifying a central role for adaptive transition policies, the aim of which is to ensure that people have the time and adaptive capabilities needed to adapt successfully to new conditions.

Damage to the frontal areas of the brain is associated with alterations in cognitive, social, and emotional regulation abilities. These neuropsychological consequences present challenges to ecologically valid assessment (difficulties in everyday life being poorly predicted by traditional neuropsychological test performance) and far transfer of rehabilitation gains to everyday life. In this thesis, the literature on 'cold' and 'hot' cognitive, social and emotional frontal functions is discussed in relation to these challenges. Gaps in the research are identified relating to 1. Associations between specific 'hot' and 'cold' cognitive processes, 2. Association of 'hot' and 'cold' cognitive processes with everyday outcomes. Four studies are presented each addressing a different aspect of these gaps: 'cold' executive and 'hot' emotion regulation abilities and peer relationships following paediatric acquired brain injury (ABI); patterns and predictors of performance of people with traumatic brain injury (TBI) on a modified gamble task compared with healthy controls; interaction between coping style and specific executive functions in association with emotional outcomes after ABI; and the effect of brief goal management training (GMT) and periodic alerts on achievement of everyday intentions following ABI. Results indicate variation in the extent to which 'hot' and 'cold' frontal functions are associated with each other and everyday outcomes. A 'frontal-contextual system' model in which performance characteristics arise from the dynamic interaction between 'hot' and 'cold' frontal systems and everyday practical and social contexts is presented as a way of understanding everyday difficulties. Application of novel methodologies that can sample the interactions between system components and are sensitive to inter-individual variability may be useful for advancing understanding of the links between frontal dysfunction and everyday life. Implications for intervention that are similarly focused on the interactions between components and facilitation of social or physiological conditions that give rise to optimal adaptation in everyday life are discussed.

Neurodegeneration, the progressive and irrevocable loss of neuronal structure, is quickly becoming an imposing health concern in a globally ageing society. While specific neurodegenerative conditions exhibit specific clinical symptoms and progressions, a common neuropathological feature is the misfolding, oligomerisation and fibrillation of certain proteins causing neuronal stress and death. Parkinson’s disease, PD, has long been characterised by the death of nerve cells focused in the substantia nigra pars compacta region of the midbrain and deposition of large protein aggregates, called Lewy Bodies, throughout the central nervous system. More recently, the protein which forms these inclusion bodies was identified as alpha synuclein, αSyn, a ubiquitous neuroprotein with no known function. Furthermore, persons with mutations in the SNCA gene, which codes for αSyn, exhibit PD progression at a far younger age with a more severe phenotype, positively linking αSyn with PD. αSyn is an intrinsically disordered protein, IDP, and generally persists as such in solution and inside bacterial and mammalian cells. However, when in contact with a lipid bilayer the protein will embed upon the surface in an amphipathic alpha helical conformation and can also aggregate, forming toxic oligomeric and fibrillar species containing significant β-sheet identity. Its function as a helical apolipoprotein and subcellular localisation to both the nucleus and synapse has led researchers to suggest that αSyn has a role synaptic transmission and release. However, knocking out the protein does not reduce viability or produce pathological abnormalities in neuronal structure. The helical form of the protein may also persist as transient, metastable helical bundles which are non-toxic and resist aggregation. While a number of studies and tools have been reported and developed to investigate the toxic oligomeric/fibrillar forms of αSyn, very little attention has been accorded to the helical conformation. This thesis will redress this balance by producing tools which will allow us to mimic the helical form of αSyn, promote the active refolding of the full-length protein using a stable, helical peptide template and produce antibodies which recognise helical αSyn specifically for use in discovery and chaperone-like refolding. In Chapter 2 a region of αSyn (14 amino acids) was identified with a unique primary sequence located within a mutation prone section of the protein. Peptide ‘stapling’ technologies were then employed using a panel of monosubstituted ‘staple’ diastereomers, to produce a highly helical portion of αSyn. Using several other protein targets particular diastereomeric ‘staple’ combinations were analysed for obvious trends in helical content. Using solution NMR, backbone refined three dimensional structures of these helical peptides were produced which showed that they were faithful structural homologues of their parent helical proteins. In Chapter 3 the drug-like properties and therapeutic potential of stable, helical αSyn peptides were investigated. Using fluorescently labelled peptide substrates, ‘stapled’ peptides were shown to be far more cell penetrant than their wild type equivalents and demonstrated that the mechanism for cellular uptake appears to be specific. Furthermore, under harsh proteolytic conditions the ‘stapled’, helical peptides were far more resistant to hydrolysis than wild type or ‘stapled’, poorly helical peptides. The ‘stapled’ peptides were also highly soluble and did not appear to aggregate in a time-dependent manner. Using ion mobility mass spectrometry, it was shown that incubation of full-length protein with the ‘stapled’, helical peptides caused a contraction in the hydrodynamic radius of the protein. However, using solution NMR no active refolding of αSyn was observed when under the same conditions. Rather small perturbations in chemical shift were apparent which did not suggest that the αSyn protein folded into a discrete structural conformation, such as an alpha helix. In Chapter 4 the stable, helical αSyn peptide was employed as a conformational model and unique antigen in antibody discovery. Immunisation with the ‘stapled’, helical αSyn peptide initially produced a pool of polyclonal antibodies with a half log specificity for the helical peptide. After bespoke affinity chromatography this was increased to three log orders of specificity. Initial immunocytochemistry did not detect any helical αSyn protein in SH-SY5Y cells. To validate the helical epitope on the full-length protein in vitro an assay based around flow cytometry of synthetic vesicle structures was developed, with their synthesis, characterisation and binding of the αSyn protein described.

This thesis aims to advance our understanding of the spatial and temporal propagation of economic shocks across the UK economy. The first chapter sets out the data I will use throughout this thesis. The second examines the role of cash-flow shocks on household and firm behaviour. I find that the reduction in monthly mortgage payments resulting from monetary policy stimulus in the autumn of 2008 significantly boosted employment in the depths of the Great Recession. The neighbourhood-level cash-flow effect varied with the proportion of fixed-rate mortgagors across the country, and its interaction with the structure of local employment markets led to regional heterogeneity in the potency of monetary policy. I find that the effect of monetary policy on overall annual employment growth varied by around 1.5 percentage points across local authorities. The third chapter explores how local demand shocks associated with house price changes affect employment growth, and whether these shocks are transmitted across firm establishment networks. I find little evidence that credit or informational frictions have played much of a role since the Great Recession. But the employment effects of these local demand shocks do extend beyond administrative boundaries and ripple short distances across the country. The fourth chapter revisits the Granular Hypothesis and its implications for the UK. I find that the UK firm-size distribution exhibits a quantitatively fat tail, but this is dominated by firms that generate revenues from customers living nearby. The size of such firms is naturally driven by the number of establishments in their network, which means they tend to have low geographical concentration. Since I find limited evidence for establishment-network transmission of local demand shocks, this suggests the propagation of idiosyncratic shocks to large firms is more likely to emanate from supply chain disruptions than from shocks to demand.

Mathematical and computational modelling provides a useful framework within which to investigate the organisation of biological tissues. As experimental biologists generate increasingly detailed descriptions of cellular behaviour, models that consider cells as discrete entities have become a common tool to study how cell-level processes affect collective dynamics, form and function at the tissue level. To date, however, models incorporating detailed biophysical descriptions of cell shape dynamics have gained little traction among the modelling community. Few model implementations are publicly available, and there often remains no comprehensive account of their method of solution, computational implementation, or analysis of parameter scaling, hindering our ability to utilise such models in practice. In addition, the quality of software underpinning such models (and academic research more widely) is coming under increasing scrutiny, with a growing recognition of the need for correct, reliable and sustainable research software tools. This thesis aims to address these needs for one such cell-based modelling approach, the immersed boundary method. We develop and analyse the immersed boundary method and provide an efficient, open source implementation for simulating cell populations. The implementation is undertaken within Chaste, an open source C++ library that allows one to easily change constitutive assumptions, and is designed and built with software best practices in mind. We explore such best practices and refine and add to the infrastructure and functionality of Chaste. We then carefully compare the immersed boundary method with the vertex model, a competing cell-based modelling approach for epithelial tissues, clearly elucidating relative strengths and weaknesses of the immersed boundary method. This furthers our understanding of the circumstances under which the immersed boundary method and similar cell-based approaches are an appropriate computational tool. Finally, to demonstrate the efficacy of the immersed boundary method, we investigate the mechanics underpinning a novel form of epithelial bending, advancing our understanding of the formation of placodes, structures of epithelial thickening in the cranial region of developing embryos. Together, the contributions in this thesis advance the use of reproducible software development approaches in cell-based modelling.

The friction of tyres on roads has been of practical importance for many years with nearly 80% of terrestrial traffic making use of rubber tyres. Tyres provide the grip required for vehicle acceleration, braking and cornering. In order for a tyre to grip on a snow covered surface friction mechanisms such as “ploughing”, (where sharp tread block edges dig into and break bonds between the snow grains) and fluid film lubrication must be considered. These are not present when a tyre interacts with tarmac. In addition metamorphism of the snow over time can result in variations of the structure and mechanical properties, this can occur rapidly especially when dealing with temperatures close to snows melting point. When full car-scale outdoor testing is carried out the snow conditions cannot be controlled and vary daily. This means the snow properties must be measured every day so that any observed variations in friction can be attributed to the tyres rather than the snow. At present the simple measurements being carried out on the snow tracks have not proved sufficient to pick up on the variations in the snow. This leads to inconsistent results: one tyre behaves differently on two different days, even though the snow was measured to be the same. This has resulted in the need for further study of the way snow variations influence the grip of winter tyres. The primary aim of this study is to identify which snow properties contribute to the friction of tyres on snow and be able to estimate the friction from measurements of snow properties. This work is the first comprehensive study to combine: multiple snow properties, microstructure characterisation, measurement of friction behaviour and different snow (both artificial and natural). In order to study the way snow affects the grip of winter tyres, methods of manufacturing artificial snow with consistent mechanical properties and microstructure are used. A method of blending ice chips (a solid state fracturing process) and compressing the resulting snow to form a test track was developed during a previous PhD carried out in our group. An alternate snow microstructure was created by using an established process of creating snow by vapour deposition. The process was simplified and downscaled, the resulting snow consisted of large dendritic grains, very different to the blended ice chips. Both snows were pressed in identical manners to create snow testing tracks. In addition, natural snow collected from the field was tested to compare with the artificial snow. In order to investigate how the variations in the snow affected the friction of tyres extensive testing was carried out in a cold room using a linear tribometer, using procedures established in previous studies. Two analytical rubber samples were used to investigate the friction, a rounded edge sample and a siped sample. Testing was carried out at -10°C at speeds of 0.01m/s, 0.1m/s and 1m/s. A significant part of this PhD involved the development of new methods and equipment which have not been used to study snow in this way before. In order to characterise mechanical properties, shear testing, compression testing and cohesion testing were carried out. To investigate snow microstructure, surface profilometry, microscopy and X-ray microtomography were used. Correlating the changes observed in snow characteristics with the changes recorded in the coefficient of friction has allowed the development of an empirical equation. This can be used to predict the coefficient of friction of a given snow based on three relatively simple snow measurements: a compression test to calculate the effective modulus, a roughness measurement to calculate the peak count density and a snow penetration test. For the first time this study allows us to use the empirical equation to estimate the relative contributions of the ploughing and surface friction mechanisms to the total friction. This allows the comparison of full car-scale test data as it is now possible to account for variations in the snow test tracks.

Rodents are the most speciose mammalian order and are represented in arboreal, semiaquatic, subterranean and terrestrial niches. To flourish in such environments, rodents must exhibit morphological traits that can reflect functions that are needed to survive. This thesis focuses on the functional morphology of digging subterranean rodents and in particular, African mole-rats (Bathyergidae). Species dependent, subterranean rodents dig using a number of different methods. This thesis concentrates on the morphological differences in the craniomandibular complex in scratch digging and chisel-tooth digging subterranean rodents. Scratch digging rodents use only their claws to remove softer soil whilst their chisel-tooth digging counterparts use their incisors in concert with their powerful masticatory muscles to remove harder soils. Chapter two looks at morphological traits associated with bite force and gape in African mole-rats (Bathyergidae). The study shows that chisel-tooth digging rodents have morphological traits that are associated with a larger bite force at wider gapes, which is probably achieved by having a temporalis with a greater mechanical advantage. Chapter three examines a selection of chisel-tooth digging, scratch digging and terrestrial rodents. It shows that the upper incisors of chisel-tooth digging rodents have a larger radius of curvature. Also, it shows that chisel-tooth digging rodent cranial shape converges in morphospace and covaries with the upper incisors, although these results were not significant when phylogeny was accounted for. Chapter four shows that, using finite element analysis, the cranium of a chisel-tooth digging mole-rat can create larger bite forces at wider gapes, compared to a scratch digging mole-rat. Using a novel method of combining geometric morphometrics with finite element analysis, this study also shows that the cranium of the chisel-tooth digging rodent deforms less, making it more efficient at performing chisel-tooth digging tasks. Overall, this thesis shows that the craniomandibular form of subterranean rodents can be strongly influenced by function. The digging method used by a subterranean rodent is therefore important to how they have evolved.

Despite recent improvements in the clinical outcomes of vital pulp treatment, existing approaches remain non-specific and unpredictable. Developing biologically-based therapies that promote pulp regeneration is critical. Epigenetic modifications of DNA and histones control cellular processes, including proliferation, mineralisation and stem cell fate, and therefore offer exciting therapeutic opportunities. Chromatin acetylation can be altered pharmacologically using histone-deacetylase-inhibitors (HDACis), which relax its structure and modulate transcription. This project investigated regenerative-associated HDACi effects in vitro on a cell-line and primary dental-pulp-cells (DPCs), using proliferation, viability, mineralisation, cell-migration, enzyme activity, high-throughput gene/protein expression and pathway analyses. HDACis increased DPC differentiation and mineralisation-associated gene/protein expression at concentrations, which did not reduce viability. Primary DPC mineralisation was promoted without altering cell viability/apoptosis, indicating a resistance to HDACi-mediated toxicity compared with cell-lines. HDACi-induced DPC reparative processes were mediated by matrix metalloproteinase (MMP) expression and activity. MMP-13 inhibition further increased mineralisation-associated events, but decreased cell-migration indicating a novel role for MMP-13 in pulpal repair. HDACi solutions released a range of previously characterised and unreported bioactive dentine matrix components, which may further supplement regenerative capability in vivo. Results demonstrate that HDACi directly stimulate DPC repair-associated events, highlighting their potential for augmenting dental materials or pulp-engineering scaffolds for regenerative endodontics.

Insects are richly instrumented with sensors used in flight control. Whilst we have good knowledge of the basic physiology of these structures, in many cases we have only a cursory understanding of the likely stimuli each sensory system receives during flight or why these systems have evolved to perform their particular function. This thesis examines the role and optimisation of arrays of cephalic flow-sensitive hairs, called trichoid sensilla, in the flight control of the desert locust (Schistocerca gregaria). Particle image velocimetry and computational fluid dynamics are used to measure and simulate the airflows being received by the fields of sensilla during flight. The visualisations reveal distinct local flow conditions over each field and how these are modulated differently by perturbations. The geometry of every individual hair is captured using both benchtop and synchrotron based X-ray micro tomography (XMT). These scans enable three-dimensional measurement of each hair's direction of curvature, which represents the direction of peak sensitivity to airflow. By combining flow stimulus and directional sensitivity data, we provide strong evidence for separate fields being tuned for distinct roles: some as step detectors for monitoring stall at extreme angles of attack, others as graded sensors for angles of attack and sideslip, and possibly for monitoring airspeed. Finally, we examine the flight response of tethered locusts to varying aerodynamic stimuli by changing the angle of sideslip and selectively covering fields of sensilla from oncoming airflow. We demonstrate a clear role for the sensilla in modifying the forces and moments generated under varying sideslip, as well as in changing the underlying natural modes of motion of the system. We reveal the axes about which the locusts are unstable under our experimental conditions, and demonstrate the necessary additional sensory feedback required to stabilise the system fully. The combined data represent a comprehensive and thorough examination of the complete cephalic flow-sensing system, providing insights applicable to both further biological study on multi-modal flight control and the use of bio-inspired arrays of sensors for manmade applications.