42337 research outputs found
Sort by
Mapping the cell cycle-dependent centrosomal interactome
No full textA series of highly regulated and precisely ordered events must occur for cells to multiply. Understanding the dynamic interactions of proteins driving this process are crucial for deciphering cell fate. The centrosome is a non-membrane-bound organelle that acts as the primary microtubule organizing centre and is believed to act as a central hub for numerous cell cycle regulators. Consistent with this view, multiple cell cycle regulatory proteins have been reported to facilitate faithful cell division processes through their centrosomal localisation. However, a detailed mechanistic understanding of how these signalling proteins regulate cell cycle transitions from the centrosome remains elusive. To better understand the regulatory signalling networks anchored on the centrosome, an endogenous proximity labelling system called miniTurbo was fused with centrosomal proteins Centrin 2, PCNT and Cep192 to detect the centrosome interactome. The cell cycle-dependent interactome of the centrosome was mapped by combining proximity biotinylation and a reversible drug-mediated cell cycle synchronisation method based on the CDK4/6 inhibitor palbociclib. Differential analysis of the centrosome interactome resulted in the identification of proteins associated with numerous biological processes, including gene expression, DNA/RNA processing, and cell cycle regulation. Functional analysis of three enriched interactors, AATF, PPP1CB, and TRIM25 was performed. Initial analysis on AATF revealed that this protein participates in the G1/S and G2/M transition when depleted using the auxin-inducible degradation system. Unfortunately contrary to the enrichment results, fluorescently tagged AATF-AID2-mNeonGreen was found to be predominantly localised on the nucleolus. Similarly, PPP1CB-AID2-mNeonGreen could not be detected on the centrosome, and its degradation showed no effects on cell cycle progression. For TRIM25-AID2-mNeonGreen, it was observed to selectively associate with the centrosome during the G2 phase, and degron mediated depletion experiments suggest that TRIM25 may be involved in G2 phase regulation. This work presents an initial study of protein dynamics at centrosomes during distinct cell cycle stages and ongoing validation of enriched interactors is anticipated to reveal additional centrosomal cell cycle regulatory proteins
The role of heat and mass transfer on the effluent generated by fire flames
No full textThe precise chemical composition of fire effluent has eluded scientists for decades. When solid materials burn, gaseous chemical species are initially generated. The production of gaseous volatile species occurs via pyrolysis, initiated by heat transfer. Pyrolysis products are subsequently oxidised by a flame to produce combustion products. These combustion products, alongside a proportion of partially oxidised pyrolysis products, are the species entrained in air forming the aforementioned effluent. It is therefore important to consider both preceding processes in order to fully understand the composition of fire effluent.
The resultant analytical problem is twofold: to be able to identify the combustion products being produced, a study must be confident in identifying the initial volatile species being generated. This generation stems from a combination of the initial material composition, the rate at which mass is being lost, the geometry of the flame and the imposed combustion environment. Increasing the oxidative environment using existing methodologies results in the mass loss rate (MLR) of a sample increasing as the fire intensifies. This elevated MLR causes the pyrolysis rate to increase, enabling a greater proportion of oxidation reactions to occur. This increased volume of exothermic oxidative processes causes the overall heat release rate (HRR) to rise. Such effects result in the surface of the sample reaching greater temperatures, resulting in thermal gradients being established within the material under investigation. These thermal gradients alter the pyrolysis pathways available, preventing comparisons between experiments conducted under differing oxidative environments being linked to a single changeable variable.
This thesis attempts to decouple the link between the MLR of a sample and the oxidative environment, thus enabling an insight into the role of heat and mass transfer on the composition of fire effluent. Initial room-scale experiments were conducted to attempt to obtain representative fire effluent for analysis. The combustion environment obtained during these room-scale experiments enabled the interaction between a descending smoke layer and the flames to be assessed, however the flow of oxidiser could not be independently adjusted without altering other variables such as the MLR, thus, an alternative approach was required.
Reducing the scale of the problem was found to enable a greater level of control over experimental variables. The modified means of controlling the Fire Propagation Apparatus (FPA) allowed the MLR of a sample to remain fixed under differing oxidative environments. A proportional, integral derivative (PID) controller was used to adjust the voltage controlling the FPA lamps based upon a live reading from the inbuilt load cell. This resulted in a means of enabling a steady state MLR whilst recording the mass and observing the sample throughout an experiment. The combination of such data enabled near real-time yields to be calculated and linked to observed flame geometries.
Furthermore, the heat flux being sent to control the MLR was recorded, enabling the quantification of various oxidative processes when used in conjunction with the effluent analysis. The use of a combination of analytical techniques enabled a greater proportion of the species generated in the effluent to be identified, providing insights into both pyrolysis and combustion processes. Such a setup enabled the effects of oxygen on pyrolysis and flaming combustion over a range of materials to be investigated in a novel manner.
Decoupling the MLR from the oxidiser has enabled the effect of flame geometry to be linked to the effluent composition for condensed-phase fuels. By demonstrating that these flames behave as diffusion flames, the fire science field can begin to move away from treating fire flames as point sources for chemical species to enter the effluent. It is hoped that the methodology developed in this thesis will continue to enable effluent composition to be linked to changing oxidative environments in a manner that successfully decouples the symbiotic relationship between the MLR and the oxidiser
A laboratory investigation of CO₂ storage by mineral carbonation
No full textAchieving geological carbon mineral storage (CMS) through engineered in-situ
mineralisation has become established as a climate mitigation technology over the
last ten years. Despite the success of various demonstration projects, uncertainties
remain over the mechanisms of carbonate formation within reservoirs, and grain scale-
processes which could aid or inhibit ongoing mineral precipitation and thus affect the
long-term viability of geological repositories. This thesis presents the results of a
campaign of laboratory experiments designed to further knowledge about the physio-
chemical phenomena which underpin the mineralisation reaction. Original research is
presented in three thematically linked journal articles.
The first paper presents the results of a suite of batch experiments conducted
over the course of several years. These investigate the effects of time, temperature,
CO₂ partial pressure (pCO₂), and fluid chemistry on carbonate precipitation in the
CO₂-water-basalt system. I show that for CO₂-water-basalt experiments at low
temperature and pCO₂ ranges (45-85 C and 0.04 – 1 bar pCO₂), carbonates do not
precipitate in experiments of up to one year’s duration. Instead, minor clay and zeolite
alteration are the only discernible mineralogical changes from high sensitivity
analytical methods. Only experiments with the addition of 0.5 M NaHCO₃ solution
display the precipitation of carbonates over measured timescales. I use these results
to deduce that formation fluid pH is a critical factor when assessing the viability of
potential CMS reservoirs.
The second paper presents the design, construction, and operational demonstration
of bespoke reservoir simulation apparatus I have developed during the PhD to
address a gap in the published literature. This compact reaction cell design allows
for the simulation of CMS reservoir environments with realistic physical conditions;
whole rock cores and fluid flow. The cell is x-ray transparent and allows the collection
of time-series microtomography (4DμCT) data which tracks physical changes in the
rock as the mineralisation reaction proceeds. I demonstrate the functionality of the
cell and include a complete set of technical drawings for other researchers to reproduce
the cell.
The third paper presents the results of a 93-day operando experiment simulating
a basalt-hosted CMS reservoir environment using the methodology presented in the
previous chapter. I document the entire lifecycle of the mineralisation reaction using
a combination 4DμCT and a host of other microanalytical techniques, investigating
the diagenesis of newly formed carbonates within the basalt sample. I show the
development of a nascent fracture network and argue that the data present a strong
case for an underlying self-sustaining process of reactive surface generation, driven by precipitation of carbonates. I show a bulk increase in sample porosity of up to 2%
and a final volume fraction of carbonate minerals of up to 1.56% over the duration of
the experiment. The 4DμCT data presented are novel in terms of both the underlying
fidelity to CMS reservoir environments, and the duration of reaction investigated.
The presented results have applied value in reservoir characterization, understanding
fundamental physical processes, and validation of subsurface models. The presented
methodologies and results offer scope for a continued programme of experimental
work to further knowledge of CMS reservoir environments
Narrating (in) transformation: contextual narratology, gender, and the dialogic interplay between narrative form and narrated content
Get PDFThis doctoral thesis explores the dialogic relationship between narrative form and narrated content through the lens of contextual narratology, focusing on gender-conscious novels that emphasise identities in transition. Specifically, my research approaches narrative form as a system of interconnected, interdependent and mutually illuminating relations (both structural and thematic), the detailed study of which can reveal gender- and queer-related significations when examined through a contextual narratological lens.
Challenging the view of formalism as either rigidly tied to structural foundations that prioritise form over content—as is often the case in classical narratology—or as restrictive or irrelevant to gender- and queer-related meanings—a stance commonly associated with gender and queer studies—I propose an alternative conception of formalism. I frame it as a methodological approach that occupies a middle ground, namely, a study of narrative form in dynamic interaction with narrated content, employing the analytical tools of narratology while also engaging with theoretical concepts from gender and queer theory.
To foreground this understanding of narrative form, I turn to the principle of transformation—a concept rooted in Structuralism but still underexplored in terms of its potential for narrative analysis and processing. By tracing its development in narratology and its more recent engagement within queer narrative theory, I argue that transformation serves as a crucial link between narrative form and narrated content, as it encapsulates the dynamic interplay between a narrative’s structural shifts and the thematic changes it conveys.
Moreover, a central aim of this research is to examine the potential role of gender as a productive element in narrative processing. I respond affirmatively to the question—first raised in the 1980s and 1990s but still largely unanswered—of whether gender can productively inform narrative analysis. To this end, I explore gender through its intratextual, formal function: not simply as a contextual factor, but as it operates in relation to structural narrative elements such as temporality, perspective, focalisation, narrative space, characterisation, and metafiction. In doing so, I underscore the importance of contextual narratology—particularly feminist, rhetorical, and queer approaches—in uncovering nuanced gendered and queer significations. Finally, I demonstrate the limitations of classical narratology in addressing such narratives, highlighting the need for more expansive and critically engaged narratological frameworks
SALL protein tetramerisation is essential for transcription factor function
No full textMammalian Spalt-like proteins (SALL1-4) are important for early embryonic
development and organ formation. These proteins feature a conserved N-terminal
domain that interacts with the NuRD co-repressor complex and several C2H2 zincfinger
clusters, of which zinc-finger cluster 4 (ZFC4) is required for binding to AT-rich
DNA. SALL1 and SALL4 are expressed in mouse embryonic stem cells (mESCs) and
are mutated in human congenital disorders, Townes-Brocks syndrome (TBS) and
Okihiro syndrome respectively. SALL proteins can form multimeric complexes whose
nature and biological function remain unclear. This research project aims to
understand the molecular mechanism and potential functions of SALL protein
multimerisation.
I defined a 33 amino acid glutamine-rich (Q-rich) region near the N-terminus of SALL4
that is necessary and sufficient for interaction with endogenous SALL1 and SALL4 in
mESCs. AlphaFold predicts this Q-rich region to form an alpha-helix and SEC-MALS
analysis on recombinant protein complexes detected a discrete and stable tetramer.
When modelled in mESCs, a missense mutation in the Q-rich region of SALL4 found
in an Okihiro syndrome patient abolished multimerisation and disrupted nuclear sublocalisation
to AT-rich pericentric heterochromatin. This resembles defects in DNA
binding due to pathogenic SALL4 mutations in the DNA binding domain ZFC4,
indicating a potential role for transcription factor multimerisation in efficient binding to
target genes. Genome-wide approaches including CUT&RUN and RNA-seq have
verified a loss of DNA binding in Q-rich mutants and consequent deregulation of target
genes. Taken together, these findings demonstrate that SALL4 multimerisation is
required for correct genome binding and transcription factor function.
Expanding from what was learned about the Q-rich region in SALL4, I investigated
the SALL1 Q-rich region which is very conserved. Modelling the Okihiro syndrome
mutation in SALL1 abolished multimerisation and SALL1 missing the Q-rich region
altogether did not co-localise with AT-rich pericentric heterochromatin, indicating
functional conservation. Furthermore, a potential dominant-negative effect of mutated
SALL1 that is found in Townes-Brocks syndrome patients and is linked to a more
severe phenotype was studied. This prevalent TBS mutation that leads to a loss of
the ZFCs but retains the Q-rich region was able to redistribute SALL4 away from its
binding sites at AT-rich pericentric heterochromatic foci in fixed and live mESCs.
Thus, disease-truncations might reduce the DNA binding ability of the heterocomplex
and exert a dominant-negative effect on other SALL proteins, contributing to the more
severe phenotype seen in patients with these mutations.
Taken together, these results show that SALL protein multimerisation is crucial for the
function of this transcription factor family and shed light on the molecular basis of
Townes-Brocks and Okihiro syndromes
Essays on strategic communication
Get PDFThis thesis consists of three chapters, each of which is self-contained and studies
a distinct theoretical setting. All chapters are linked by a common focus on
strategic communication, where private information is communicated strategically.
CHAPTER 1: OPTIMAL BONUS STRUCTURE UNDER STRATEGIC COMMUNICATION
Bonus structures are a common tool used by organizations to incentivize workers’
efforts. Halac, Lipnowski, and Rappoport (2021) assumes workers cannot
observe each other’s contracts and derives the optimal bonus structure that
uniquely implements work. In reality, however, workers can strategically communicate
with each other before exerting effort. I study the optimal bonus
structure that uniquely implements work when communication is allowed. I
model the interaction as a two-stage game: a communication stage where workers
communicate with each other, and an action stage where workers choose
whether to work. I show that the optimal bonus structure is robust to communication
that takes the form of Bayesian persuasion and cheap talk, in the
sense that the unique equilibrium involves workers sharing no information in
the communication stage and always working in the action stage.
CHAPTER 2: EVALUATING EXPERTS UNDER IMAGE CONCERNS
An expert with image concern wishes to persuade others that he holds a certain
opinion, which could result in the expert failing to report his opinion truthfully.
Can we learn about the expert’s true opinion under such image concern
by evaluating him appropriately? I characterize robust evaluation rules that
implement truth-telling equilibrium under any level of image concern. I further
derive robust evaluation rules that minimize the cost of incentivizing truthtelling
and show that such evaluation rules always provide a higher material
payoff to a dissenting report.
CHAPTER 3: THE MARKET FOR EXPERT ENDORSEMENT
When experts are hired as tools of persuasion, how are their incentives to
reveal information affected? This paper studies a three-party model (expert,
manager, public) where an expert, either informed or uninformed, is hired by
a manager to persuade the public about an uncertain state. The expert wish
to be perceived as informed. In a baseline setting, I show that the reputation
concern lead to the expert providing endorsement inconsistent with his private
beliefs. In the second and third settings. I show that information control
power is only beneficial to the manager when combined with commitment
power, in which case, the uninformed expert becomes more valuable than
the informed expert to the manager in terms of increasing the probability
of successful persuasion
Making Systems Answer: A Practitioner's Handbook for Trustworthy Autonomous Systems
Get PDF36 pagesThis handbook offers a practical guide for organisations wishing to deploy autonomous systems in more trustworthy, safe and responsible ways. It draws upon EPSRC-funded research conducted at the University of Edinburgh from 2022-2024 on responsibility for trustworthy autonomous systems, understood through the concept of 'answerability.' This handbook offers an introduction to the problem of responsibility for trustworthy autonomous systems, the benefits and regulatory relevance of an answerability-based approach, a lay summary of the underlying empirical research on answerability, case studies, and tools and best practices for developing, deploying and regulating autonomous systems with answerability at the forefront
SPIKA: an energy-efficient hybrid CMOS-RRAM compute-in-memory macro for machine learning
No full textThe deployment of neural networks (NNs) in machine learning (ML) applications such as computer vision, speech recognition and natural language processing has grown exponentially in the past few decades. The biggest challenge in implementing such algorithms is the constant data movement between the compute units and memory units. Today’s computing systems, primarily built based on the von Neumann architecture where data must be moved to a processing unit, have shown inefficiency in implementing ML algorithms. The speed and energy associated with this bottleneck present a key performance concern for a range of applications in artificial intelligence (AI) workloads. Another key challenge is that NNs carry out copious calculations of Multiply and Accumulate (MAC) operations which require high-performance GPUs, consuming a great amount of power. Therefore, there is an important need to improve computing efficiency in terms of both energy and latency. Innovation in new computing architectures is expected to play a major role in the future of ML hardware.
Non-volatile compute-in-memory (nvCIM) technology has recently shown promising results in addressing the data movement and multiply-and-accumulate (MAC) bottlenecks in machine learning algorithms by enabling parallel analogue vector-matrix multiplication (VMM) operations directly within memory arrays. By executing certain computational tasks within the memory itself, nvCIM provides an efficient alternative to traditional computing approaches. Specifically, nvCIM based on Resistive Random Access Memory (RRAM) has garnered attention due to its use of Ohm’s law for multiplication and Kirchhoff’s law for accumulation, allowing RRAM arrays to perform parallel in-memory MAC operations with significantly improved throughput and energy efficiency over digital computing methods. RRAM cells are used to carry weights of the neural network due to their low read voltage, ability to achieve multiple states per cell and dense structure.
In this research work, I present SPIKA, a novel energy-efficient RRAM-nvCIM chip designed for accelerating machine learning workloads. The main aim of this PhD project is to accelerate ML and ANN applications at the maximum possible power efficiency. The key innovation of SPIKA lies in its ability to efficiently transfer input signals to output signals with minimal overhead. The analogue computation is performed in the time domain, with the dot product accumulated on a switched capacitor, eliminating the need for high-resolution, power-intensive data converters. Ultimately, the key pillar of SPIKA is that it leverages the low-resolution niche it addresses to allow each domain to play to its strengths whilst using simple and efficient domain converters. This makes for a highly functional and simultaneously energy and area-efficient implementation.
The SPIKA chip has been fabricated using commercial 180nm technology and experimentally validated post-silicon. The core block features a 64x128 memory crossbar and utilizes 4-bit input, ternary weight, and 5-bit output resolutions. The results indicate a remarkable performance of SPIKA chip with a peak throughput of 1092 GOPS and energy efficiency of 195 TOPS/W. Compared to state-of-the-art solutions, the SPIKA core exhibits a significant energy efficiency improvement, ranging from 2.15x to 390x. For experimental demonstration, a neural network trained on the Modified National Institute of Standards and Technology (MNIST) database was implemented on the SPIKA chip. Results show a minimal 3\% loss in classification accuracy compared to the software baseline with 32-bit resolution, using the same network size and ternary quantized weights.
Furthermore, an 8-core SPIKA system, each core featuring a 64×128 array, is proposed to extend the architecture. The system-level architecture introduces minimal overhead on the SPIKA core by incorporating a streamlined switching mechanism within each core, enabling efficient analogue aggregation and inter-core communication without the need for additional circuitry. Circuit-level simulations demonstrate the SPIKA system's superior performance, achieving a peak normalized throughput of 8.736 TOPS and energy efficiency of 312 TOPS/W, demonstrating competitive performance even against designs on more advanced technology nodes
The role of alternatives to primary aggregates in reducing emissions from the construction sector
Get PDFScotland's construction industry relies heavily on traditional primary aggregates. Lower carbon alternatives such as recycled concrete and incineration bottom ash aggregates are gaining traction. This study has found that that alternatives to primary aggregates can reduce greenhouse gas emissions significantly, with local sourcing further amplifying these benefits. However, logistical and supply chain challenges may limit these benefits when transportation distances exceed certain thresholds
Additive engineering mechanisms for antimony chalcogenide solar cells
Get PDFAntimony chalcogenide solar cells are an attractive thin film solar technology with a tuneable bandgap, high inherent stability and a large absorption coefficient. They are often solution processed, allowing for the easy employment of chemical additives.
Though many additives have already been utilised, the underlying chemical mechanisms are often poorly understood. The formation of unwanted side phases, most notably Sb₂O₃, are also largely not understood. Here, the chemical mechanism of an additive of proven efficacy, EDTA, is investigated through the use of NMR spectroscopy and a solution-based chemical aggregation test, as well as the use of various techniques including powder-XRD, SEM and Raman spectroscopy on Sb₂O₃ films formed using EDTA. These tests demonstrated that EDTA can control the deposition of Sb₂O₃ films, while suppressing the formation of the undesired Sb₂O₃ side-phase. The solution-based chemical aggregation test was further developed into a screening process to find additives which rival and exceed EDTA. After the discovery of successful additives using this screening process, a chemical mechanism for the suppression of Sb₂O₃ formation was found by correlating improvement of solar cell performance with the pH of the additive. The mechanism proposed in the literature for the control of film deposition by EDTA was that through the sequestration of Sb³⁺, EDTA was effectively changing the deposition mechanism from per-nanoparticle to per-ion deposition, thus forming more compact and large crystal grains. By relating the capability of an additive to bind Sb³⁺ to its relevant solar cell performance, this mechanism is further evidenced and expanded. The deeper understanding of these additive mechanisms afforded by this investigation will allow for more focussed and informed development of additives for efficient antimony chalcogenide solar cells in the future, particularly those using a benign and abundant TiO₂ electron transport layer
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!