42337 research outputs found
Sort by
Transmission mechanisms and persistence of the Great Depression in Italy
No full textThe first chapter – Labour Under Fascism – analyses the labour market reforms of
the Fascist regime in the late 1920s. It studies the evolution of wages, hours worked
and employment in 1930s, uncovering the behaviour of wages relative to other labour
market outcomes. It tests the hypothesis that Fascist labour market reforms increased
the degree of real wage rigidity, leading to the observed falls in employment and hours
worked. The chapter collects quantitative and qualitative evidence to determine that
wages were leading indicators of labour supply and that the 1927 reforms were associated
with a rise in estimated real wage rigidity.
The second chapter – Shielded to Shrink – provides a measure of Italy’s trade barriers
during the Great Depression, evaluating the direct losses incurred from its protectionist
stance. This chapter proposes the relationship between high trade barriers, firms’
market power and firm growth as a determining factor for the proliferation of small
businesses and the observed slowdown in industrial and aggregate total factor productivity
in 1930s Italy. Accordingly, we show that trade policies promoted industrial
concentration, stifling competition and contributing to the persistence of the Great
Depression in Italy.
The third chapter – The Battle for Credit – details the government’s rescues and
its recurrent reliance on domestic capital markets to meet its financing needs. It
shows that the price of credit on Italy’s financial markets continued to rise throughout
the 1930s, providing a framework that explains how government’s fiscal deficits could
divert resources away from the private sector. Ultimately, this chapter proposes that
crowding-out effects affected recovery by squeezing private investment and the rising
cost of credit on financial markets. We propose this mechanism to be a persistence
channel of the Great Depression in Italy
Towards robust and generalisable natural language predicate inference
Get PDFNatural language predicate inference is an important task of natural language processing,
which bears particular significance in identifying supporting material for answering
questions in the open domain. Given a premise and a hypothesis assertion, both
concerning the same entities as arguments, natural language predicate inference aims
to determine whether the hypothesis is entailed by the premise, i.e. whether the hypothesis
can be concluded from the premise.
In the task of predicate inference, directionality is a key characteristic, separating
it from symmetric semantic similarity. Directionality is the property whereby the entailment relation between a pair of assertions holds in one direction but not both. To
capture this property with an example, “John got his groceries at Tesco” entails “John
went to Tesco”; however, “John went to Tesco” does not entail “John got his groceries
at Tesco”. Directionality poses a unique challenge to predicate inference methods,
especially those not making a sufficient distinction between entailment and similarity.
Prior work has developed various approaches to natural language predicate inference.
These approaches include: 1) Entailment graphs, a type of unsupervised symbolic
methods, which are induced from predicate-argument mentions in vast natural
language corpora; 2) fine-tuned Small Language Models, a type of supervised neural
classifiers, trained on inference datasets; 3) few-shot generative interaction with Large
Language Models, another unsupervised approach whose importance has been rapidly
rising throughout the landscape of NLP and beyond.
In this thesis, we explore the development of robust and generalisable predicate inference
methods based on the above approaches. We start by generalizing entailment
graphs to other languages than English. We demonstrate that strong entailment graphs
can be built from Chinese corpora and reach strong predicate inference performance
which is comparable to their English counterparts. Additionally, between the Chinese
and English entailment graphs, we observe a mutual cross-lingual complementarity,
where a simple ensemble of mono-lingual entailment graphs elicits substantially
stronger performance than any individual graph.
On the other hand, (small) language models, with their unified representation spaces
for arbitrary text sequences, have become a common component in approaches to various
classification tasks. We hypothesized that small language models can also be
trained to perform predicate inference, and that multilingual language models would
also benefit from the cross-lingual complementarity and exhibit improved performance
when trained on multilingual inference datasets.
However, through our analysis of fine-tuned small language models, we find little
evidence of capability for the predicate inference task. Especially with directional
entailments, fine-tuned language models overfit to dataset-specific artefacts that infest
the inference datasets. On an extrinsic open-domain QA task, we observe that finetuned
LMs exhibit weak predicate inference performance in both English and Chinese,
where the performance is inferior to simple symbolic entailment graphs, despite the
latter being sparse in coverage.
With the rise of Large Language Models, which exhibit strong language understanding
performance in general and few-shot learning capability, many NLP tasks are
now considered solved by prompting LLMs with few-shot examples to generate the
predictions. Since the few-shot in-context learning paradigm is independent of training
datasets and the artefacts in them, we are able to use this setup to investigate the
LLMs’ predicate inference capability. Through controlled experiments, we observe
positive results overall, where few-shot LLMs yield non-trivial performance; on the
other hand, we also observe worrying statistical biases in LLM answer generation. We
identify prominent types of such biases, including an attestation bias and a relative
frequency bias. We argue that these biases must be alleviated before LLMs can be
trusted to perform robust natural language predicate inference in context.
Overall, we develop and critically compare approaches to predicate inference, and
advocate for a hybrid approach between efficient symbolic entailment graphs and versatile
neural LLMs
The construction of morisco identity in the Spanish literary imaginary (1492-1614)
No full textThe expulsion of the moriscos from Spain between 1609 and 1614 marked the end of more than a century-long struggle for the New Christian population to be accepted within Christian Spanish society. From the moment they were forcibly baptised between 1499 and 1526 until they were ordered to leave the Peninsula, the moriscos faced constant and intense scrutiny from the church, crown, and their Old Christian neighbours regarding the sincerity of their faith. Through the expulsion orders that denounced them as heretics and apostates, the moriscos were treated as a single nation that was inherently Muslim, irrespective of their geographical origins, class, gender, or degree of assimilation within society. Over the course of the sixteenth century, the moriscos’ perceived intractable Otherness would be interpreted through two lenses: cultural and genealogical difference. Their preservation of cultural practices linked to their Muslim ancestors, as well as the refashioning of the limpieza de sangre statutes by official authorities, would cast the moriscos as an internal Other within Spain.
This study posits that the perception of the moriscos as an internal Other is a form of early modern race-making. Through the lenses of cultural and genealogical difference, what we see is the racialisation of religion, presenting the moriscos as irrevocably Muslim in spite of their conversion to Christianity. In the absence of phenotypic differences like skin colour, the moriscos’ racial Otherness was “identified” through other “visible,” or affected, differences, such as clothing and the speaking of Arabic, and the “invisible” difference of “impure” lineage. The study will reveal how the conceptualisation of the morisco figure in early modern Spanish literature centres around four rhetorics of race-making, which each serve to uncover or confirm the moriscos’ inherent Otherness through: their cultural differences; the changing relationship between the labels moro and morisco; the notion of limpieza de sangre; and the idea of the nación morisca. The study will explore the manifestation and perpetuation of these four rhetorics across a number of key literary discourses, including the novela morisca, romancero morisco, historical chronicles, drama, and prose fiction. It will also highlight how the image of the New Christian population in literary discourses revolves around three pivotal moments in the moriscos’ history: the fall of Granada, marking the end of Islamic rule within Spain (1492); the War of the Alpujarras (1568-1571), and the morisco expulsion (1609-1614).
Starting with the novela morisca genre in Chapter One, I will explore how the characterisation of the Granadan noble moros provides the foundations for the image of the moriscos as innately tied to their Muslim ancestry through their cultural practices. The importance of Arabic and morisco clothing in the collective imaginary is reinforced in the romancero morisco, a style of ballad whose resurgence can be attributed to the popularity of the novela morisca. The discord between the idealised world depicted in the novela morisca and the reality of the moriscos becomes apparent in Chapter Two, which details the events of the War of the Alpujarras. Looking at three major chronicles, I will explore how terminology became an important tool in defining the morisco as a religious Other, aligning the New Christian rebels with their ancestral faith through the renovation of the term moro. Chapter Three explores how the events of the civil war had serious repercussions on the collective image of the moriscos, with them all perceived as dangerous thereafter. In Lope de Vega’s dramatic verse, we see the potential threat that the moriscos posed to Spain emerge in plots that recognise the moriscos’ lack of fixed physical differences. Through the theme of passing, earlier rhetorics of race-making begin to be questioned, with Lope de Vega highlighting the performative nature of the moriscos’ cultural differences that are at the centre of Chapter One. Furthermore, within his drama, I will consider Lope de Vega’s focus on the third rhetoric of race-making, one which is based on the moriscos’ alleged genealogical difference: the concept of blood purity. Finally, Chapter Four will examine the representation of the nación morisca within Cervantes’s prose fiction. Written around the time of the expulsion, the collective image of the moriscos is one influenced by polemical writings that sought to characterise the moriscos as inherently incapable of following the Christian faith in order to justify the expulsion. And yet, Cervantes contradicts this essentialising image of the moriscos through individual characters who profess to be “true” Christians. With these characters, Cervantes challenges the notion that the moriscos were a homogenous group, instead revealing the heterogenous nature of the New Christian population. What is more, it is in Cervantes’s prose works that earlier rhetorics of race-making are once again challenged or dismantled.
By tracing the construction of morisco identity across various discourses throughout the sixteenth and early seventeenth centuries, this study reveals how these four rhetorics of race-making are embedded within the image of the morisco as a racialised, religious Other within Spain. And yet, while each author perpetuates the ideas of the church and crown that the moriscos were a distinct nation or race through these rhetorics, within each discourse we also see instances where these ideas are contested. The result is that, in engaging with these rhetorics of race-making to construct the morisco as a racialised, religious Other within Spain, these authors simultaneously assert and destabilise the essentialised image of the New Christian population, revealing the fallibility and artificiality of these rhetorics that sought to cast the moriscos as inherently different. Nevertheless, while these authors recognise that the moriscos were not all the same, nor were they inherently different from Old Christians, the potential danger they posed to Spain was a concern these authors could not shake. Consequently, the prevailing image of the moriscos is one shrouded in uncertainty. The authors’ repeated engagement with these essentialising rhetorics of race-making would subject the New Christians to perpetual suspicion, and would continue to influence the image of the moriscos within Spain’s collective imaginary long after their expulsion
Fragility, Institutions, and Fiscal Policy – A Critical Review
Get PDFThis policy brief examines fiscal governance in fragile and conflict-affected settings (FCAS) and its role in shaping institutions. Effective engagement requires moving beyond elite-focused analyses to consider broader societal and institutional impacts.
Fiscal governance interacts with security, justice, and service delivery institutions. Policies must account for historical governance structures, societal norms, and contemporary fiscal behaviors. Reform efforts should foster trust between citizens and the state by supporting inclusive state-society bargaining.
Realistic timelines for reform are essential, as rapid implementation risks creating policies that mimic stable states without achieving functionality. As aid budgets shrink, alternative investment sources, including private sector engagement, must be approached with sensitivity to institutional stability and conflict dynamics.
This brief calls for a holistic, politically aware approach to fiscal governance, urging international actors to bridge gaps between fiscal policy, governance, and peacebuilding. A deeper understanding of the institutional dimensions of fiscal practices is crucial for fostering long-term stability in FCAS
Climate impacts on species’ distributions in space and time
No full textAnthropogenic climate change is having major impacts on biodiversity, including driving shifts in species’ geographic distributions (occupancy and abundance). As a result, a major focus of contemporary eco-evolutionary research is on understanding how species’ distributions are changing, and predicting how they may continue to change in the future. To generate projections of species’ distributions under future climate scenarios, researchers often identify relationships between climate variables and species’ occupancy or abundance over geographic space and then project these spatial relationships over time. This ‘space-for-time substitution’ (SFTS) approach is very widely applied, but rests on key assumptions. In this thesis, I first conducted an extensive review of the use of the SFTS approach for predicting eco-evolutionary responses to climate change. I then used both empirical and simulated spatiotemporal abundance data to test the reliability of SFTS for projecting changes in species’ distributions under future climate change, as well as exploring the insights that can be gleaned from spatiotemporal abundance data.
After setting out the broad context of climate change and the potential utility of SFTS (Chapter 1), I then reviewed how SFTS are used in climate change ecology and evolution (Chapter 2), identifying how the approach is applied and its limitations. This review revealed that SFTS are widely used across different subfields (focusing on genotypes, phenotypes, species’ distributions and ecological communities) and that, while distinct disciplines appear to have progressed largely in isolation, they have arrived at broadly comparable methods. I also found that across all subfields, the SFTS approach relies on common but undertested assumptions relating to the identification of causal climate-distribution relationships in space and the transferability of these relationships over time (i.e. whether relationships are equivalent in space and time). Therefore, I arrived at the recommendations that there should be a focus on testing these key assumptions and that greater communication between different subfields would be of great benefit for the SFTS approach.
Having identified that there is a limited understanding of when spatial climate-distribution relationships can be reliably transferred over time, I then tested this using long-term UK-wide butterfly abundance data from the UK Butterfly Monitoring Scheme. First, in Chapter 3, I used the Orange-tip butterfly as a case study to develop an approach for comparing spatial and temporal climate-distribution relationships. Specifically, I first identified the seasonal periods over which temperature and precipitation variables best explained variation in Orange-tip abundance. Then, using these climate predictors, I decomposed climate variation into its spatial and temporal components, which allowed comparison of the spatial and temporal effects of temperature and precipitation variables on the Orange-tip’s occupancy and abundance across the species’ UK range. The aim of this work was to identify whether the temporal relationships between climate variables and each of abundance and occupancy were consistent with causality (i.e. non-zero and in the same direction) and equivalent to those observed over space, as is assumed when making SFTS projections. This revealed complex patterns in the level of equivalence between the spatial and temporal effects of temperature on the Orange-tip butterfly’s distribution, with variation both across the species’ range and between occupancy versus abundance. Conversely, precipitation appeared to not have a causal effect on the Orange-tip butterfly’s distribution, with little effect on occupancy or abundance over time. This emphasises the importance of identifying causal relationships before making projections; if the spatial precipitation-distribution relationships were assumed to be causal, they would generate misleading projections over time. Overall, this chapter demonstrates that the reliability of SFTS projections can be highly context-dependent, even for a single species, and this highlights the need for further exploration of when spatial relationships can be transferred over time.
In Chapter 4 I then extended this approach to consider seven univoltine butterfly species in the UK, with the aim of exploring the extent of variation in level of equivalence between spatial and temporal temperature-distribution relationships, both among and within species. Applying the methodology from Chapter 3 to multiple species revealed that both the causality of temperature-distribution relationships and the equivalence between relationships in space and time was highly variable across species, distribution metrics (occupancy versus abundance), and within a species’ climate range. The finding that temperature-distribution relationships for many species are unlikely to be causal – even when considering the seasonal periods of temperature that exhibited the strongest correlations with each species’ abundance – suggests that SFTS projections of species’ distributions under climate change may often be using non-causal relationships. In addition, even where relationships were consistent with causality, the slopes of the temporal temperature-distribution relationships were generally shallower than the local spatial slopes. This suggests that, over shorter timescales, species’ occupancies and abundances could not respond rapidly enough to track the temperature-distribution relationships observed over space and so SFTS projections may overestimate the magnitude of change in the near-term. Overall, these results add to Chapter 3 in emphasising the need to generate an understanding of the scenarios, including timescales, under which SFTS projections will be most reliable.
Finally, in Chapter 5, I aimed to address the dearth of theory relating to the application of SFTS that was revealed in my review of the field (Chapter 2). To achieve this, I used individual-based simulations, incorporating adaptation and gene flow in a stepping-stone model, to examine the patterns of spatial and temporal temperature-abundance relationships that are expected under a range of climate change, evolutionary and ecological scenarios. Spatial relationships between temperature and abundance were found to be humped in the presence of temperature-determined carrying capacities and humped with a plateaued peak when carrying capacities were constant across sites and years. In general, the slopes of the temporal temperature-abundance relationships were shallower than the local spatial slopes, with spatial and temporal slopes most similar under a weaker local adaptation scenario. This suggests that SFTS may often overestimate the magnitude of population responses to temperature changes over time and highlights the importance of considering the rates of processes operating in space and time. Additionally, where local adaptation was stronger, temperature-abundance relationships over time were found to be humped, suggesting that local adaptation leaves a discernible imprint on spatiotemporal abundance data and potentially offering an alternative to logistically challenging experimental approaches for detecting local adaptation. These results also suggest that local adaptation may undermine space-for-time projections of abundance changes in the short term, before populations have had time to adapt to new conditions. A key next step is to build on this work and test the utility of this approach in real world, multi-driver settings.
Taken together, my thesis demonstrates the risks of relying on SFTS for projecting species’ responses to climate change, and emphasises the need for further validation across different study systems and timescales. Specifically, I demonstrated how both the causality of climate-distribution relationships and the equivalence between spatial and temporal relationships can vary both across species and within an individual species’ range. Additionally, I found that temporal climate-distribution relationships were generally shallower than those seen in space, suggesting that responses in time may often lag behind climate change. This emphasises the importance of understanding the speed of processes operating over space and time, and how this affects the timescales over which SFTS projections are most reliable. Importantly, my empirical and simulation work also identified that spatiotemporal abundance data has been largely underutilised for testing the key assumptions of SFTS, with my simulations (Chapter 5) demonstrating that there may be unrealised potential for this type of data to detect local adaptation of populations to specific environmental drivers. Overall, my thesis demonstrated that considering spatial and temporal environmental effects together provides an opportunity to gain insights into the processes driving abundance trends and to improve the robustness of SFTS projections
Parallelizing graph computation with automated vectorization
No full textGraph computations have found widespread use in social network analysis, bioinformatics,
and web search. Applications often need to evaluate the same graph query multiple
times over the same data graph, starting from different source vertices, referred to as
multi-instance processing (MIP). There are mainly two approaches to MIP. The first
approach is to use highly optimized multi-instance graph algorithms that interleave the
evaluation of multiple query instances to exploit computation sharing across instances.
These multi-instance algorithms are efficient but challenging to implement. The other
approach is to use general-purpose graph processing frameworks and obtain answers to
multiple query instances through serial or batch evaluation. These frameworks are easy
to program but shown to be significantly less efficient than multi-instance algorithms.
With these two existing approaches, users have to choose between efficiency and ease of
programming. In response to the challenge, this thesis presents a systematic approach
to get the best of both worlds.
In the first part of this thesis, we present MITra, a framework for composing Multi-
Instance graph Traversal algorithms that traverse from multiple source vertices simultaneously
over a single thread. Underlying MITra is a frontier-ranking model, which
provides an abstraction for graph algorithms, separating traversal logic from computation
logic. Based on this model, MITra offers an easy-to-use programming interface.
MITra enables user to compose multi-instance algorithms by programming computation
logic in a dedicated edge function following textbook algorithms, and choosing
traversal logic via frontier-ranking configuration. On the backend, MITra synthesizes
and executes the multi-instance algorithm by automatically organizing vertices into
frontiers based on their numeric rank values, automatically sharing computation across
instances and benefiting from SIMD vectorization. We further showcase the ease of use,
expressiveness, and efficiency of MITra by developing a plug-and-play web demo. In
addition, we extend MITra to take advantage of multi-core parallelization, and evaluate
the performance of MITra through extensive experiments.
The second part of this thesis presents AutoMI, a framework for automatically converting
vertex-centric graph algorithms into their vectorized multi-instance versions. A
well-developed multi-instance algorithm runs significantly faster than traditional serial
and batch evaluation, however, its design and implementation are notoriously challenging.
AutoMI relieves the burden of writing delicate multi-instance algorithms from
developers and achieves superior performance through vectorization. In addition, we
propose TrackFree optimization in AutoMI, yielding simpler and more efficient multiiii
instance algorithm implementation. To aid the decision of whether to use TrackFree
in AutoMI, we develop an algebraic characterization. AutoMI targets vertex-centric
algorithms written in the GAS (Gather-Apply-Scatter) programming model, as promoted
by major distributed graph processing frameworks. We implement AutoMI and
demonstrate its performance advantage through extensive experiments on real-life and
synthetic data graphs.
Putting together, MITra and AutoMI provide a systematic approach to easily program
multi-instance graph algorithms and achieve high performance through automated
and effective vectorization
By the powers combined: the design and study of artificial metalloenzymes as photocatalysts
No full textPhotocatalysis is a powerful technique allowing the synthesis of products not easily accessible via traditional thermal chemistry. Transition metal complexes represent a major class of photocatalysts, with cyclometallated complexes of iridium(III) and ruthenium(II) being particularly prominent. However, the field of photocatalysis suffers from the difficulties experienced in controlling the enantioselectivity of the final product. In nature, enzymes catalyze reactions yielding products with extreme enantioselectivities. However, their scope is limited to reactions found in nature and can only utilize metals which are bio-available and natural co-factors (or their analogues) of the enzymes. This hampers the range of chemistry that can be achieved. Artificial metalloenzymes (ArM) aim to bridge the gap between the two catalysis regimes, by combining novel metals with proteins, to enable new-to-nature reactions with the advantage of increased enantioselectivity.
In the present work, two distinct methods were used to create novel photocatalytic ArMs and investigate their photophysical characteristics; introduction of an unnatural amino acid that capable of chelating Ir(III) into a naturally occurring protein and using coiled-coil peptide containing bipyridine sidechains which was designed de-novo to allow the binding of Ru(II).
The first strategy relied on the introduction of the unnatural amino acid bipyridine alanine (BpyAla) containing a 2,2’-bipyridine (bpy) side chain to the human Sterol Carrier Protein (SCP-2L). The bipyridine side chain can act as a binding site for the Ir(III) ion. Initially, 4 residues (V83, A100, Q111 and M112) on the protein scaffold were mutated to BpyAla and [Ir(ppy)2(bpy)]+ (where ppy = 2-phenylpyridine) bound on those sites. This library of 4 ArMs was studied to determine their photophysical properties, particularly the absorbance, fluorescence, quantum yields and excited state lifetimes, and limited structural study was carried out via circular dichroism spectroscopy. All the Ir-ArM variants showed a blue shifted fluorescence spectra compared to free [Ir(ppy)₂(bpy)]⁺. Further, all 4 ArMs showed an excited state lifetime more than twice as long as the free complex, and as high as 632 ns in the case of A100Bpy-Ir(ppy)₂ variant (vs 43 ns for the free complex). The incorporation of the Ir(III) center into the protein also resulted in up to 15x higher quantum yields vs free complex in aqueous solutions. Having formed and studied the photophysical properties of the Ir-ArM, a photocatalytic reaction was attempted (the homodimerization of chalcone) using the designed metalloenzyme as the catalyst. While the observed enantioselectivity was not very high with any of the modified proteins, the N-gly-M112Bpy-Ir(ppy)2 variant gave the highest e.e. (10%) of those tested. While attempts were made to introduce different Ir(III) complexes with various substituted phenylpyridine ligands to the protein, they proved unsuccessful.
Similar to enzymes found in nature, peptides can also be used to confer stereoselectivity in photocatalytic reactions. The second method of synthesising ArMs discussed in the following pages starts with de-novo designed peptides where the desired structural features can be incorporated at the design stage. Collaborators at the University of Bristol designed a hexameric coiled-coil peptide of the form A₃B₃ containing a hydrophobic tunnel, in which the peptide strand B contains BpyAla. Hydrophobic interactions pre-arrange the bipyridine residues in a manner which allows them to chelate Ru(II), to form an octahedral [Ru(bpy)₃]²⁺ complex, which is a well-known photocatalyst. The formation of the peptide-Ru ArM was followed by photophysical studies. When only peptide B was used to form [Ru(bpy)₃]²⁺ the excited state did not change appreciably form that observed for the free [Ru(bpy)₃]²⁺ complex in an aqueous solution. However, the coiled-coil system (PeptideA)₃(PeptideB)₃Ru(bpy)₃ showed a 7% longer lifetime vs [Ru(bpy)₃]²⁺. This system also showed drastically reduced quantum yield (0.6%) vs the free [Ru(bpy)₃]²⁺ complex (3.9%)
Policy development for diet and climate
No full textThe Climate Change Committee’s 2023 Report to the Scottish Parliament called for stronger action on food system emissions. Policy interventions need to address the environmental impacts of food production and consumption while ensuring dietary improvements and economic sustainability.
This report assesses Scotland’s diet and climate policy landscape, identifying areas for policy development and providing recommendations to support the Scottish Government’s climate, public health, and food security goals going forward.
The study combined desk-based research, stakeholder engagement and categorisation using a PESTLE (Political, Economic, Social, Technological, Legal, and Environmental) framework
Understanding the ‘China Factor’ in the Russia-Ukraine War
No full textThis paper examines the evolution of China-Ukraine relations from 1991 to April 2025, focusing on the strategic, economic, and geopolitical implications of China’s role in the context of Russia’s full-scale invasion of Ukraine. Traditionally centered on trade, agriculture, and strategic industries, China’s engagement with Ukraine is part of a broader effort to reshape the international order and expand Beijing’s global influence. Since 2022, China has pursued a dual-track policy—maintaining close ties with Moscow while seeking to preserve relations with Europe—leading to increasing skepticism in Kyiv.
Ukraine’s early hopes for China’s constructive role in conflict resolution have faded amid evidence of economic and military support for Russia. Beijing’s diplomatic positioning, especially in multilateral forums, seeks to avoid direct criticism while appealing to the Global South. This reflects China’s competing interests: countering US dominance, encouraging European strategic autonomy, and bolstering its leadership claim among developing nations.
The report argues that China’s ambiguous stance complicates Ukraine’s foreign policy and post-war planning. Key recommendations include reducing dependence on Chinese investment, strengthening European partnerships, and ensuring European leadership in any future peacekeeping or reconstruction efforts. Understanding the “China factor” is essential for shaping a stable and balanced security environment in Ukraine and Europe
Exploring genetic alterations and protein marker expression in cancer: investigating ARID1A and SMARCA4 knock-outs in colorectal cancer and evaluating histological markers for targeted therapy in ovarian cancer
Get PDFThis dissertation presents two distinct yet connected research projects focused on the
alteration and evaluation of DNA and protein expression in common cancers, both of
which are subjects of ongoing research into their mechanisms and potential therapeutic
advancements. Colorectal cancer, one of the most prevalent and deadly cancers
globally involves metastasis and therapy resistance driven largely by tumour plasticity,
which is affected by changes in the expression of epigenetic modifying proteins.
Ovarian cancer, on the other hand, is common among women and exhibits various
histotypes that differ in their origin, tumour microenvironment, and genetic and
epigenetic expressions. Both cancers stand to benefit from further study to facilitate the
development of new treatments.
The first project focuses on producing SWI/SNF CRISPR/Cas9 knockouts in a mouse
colon organoid line (Apcfl/fl;KrasLSL-G12D/+;Trp53fl/fl). Knockouts were validated
using qPCR and Western blot analysis. Incomplete knockouts were produced,
indicating the need for further evaluation using DNA sequencing. Additional knockout
experiments are recommended, utilising the existing protocol with different clones
picked from the lentiviral transduction pool.
The second project involved the use of TMAs (tumour microarrays) taken from patients
with different ovarian cancer histotypes. These TMAs were sectioned and stained for
membrane proteins that are potential targets for antibody-drug conjugate (ADC)
therapy. The TMAs were then scored by two independent assessors for the expression
of these targets, providing a foundation for future studies exploring the suitability of ADC
treatments in different patient demographics and ovarian cancer histotypes, as well as
the role of target expression levels.
Though these two projects investigate different types of cancer, together they provide a
broad foundation in research methods and the diverse skills required for cancer
research. Both projects contribute valuable groundwork for future studies on the
mechanisms and treatment of colorectal and ovarian cancers
Access Repository Dashboard
Do you manage Edinburgh Research Archive? Access insider analytics, issue reports and manage access to outputs from your repository in the CORE Repository Dashboard!