Phagocytosis, the engulfment and elimination of particles by cells, is a fundamental component of the innate immune response. Macrophages (phagocytosing white blood cells) phagocytose bacteria to clear infections. S. Aureus bacteria which have been phagocytosed by macrophages have been observed to survive, replicate, and escape the host cell, killing it. Mechanisms of why this occurs have been identified, such as reduced acidification of the phagosome (compartment into which the bacteria is engulfed), however why this occurs in particular individual macrophages but not others is not clear. Machine learning will be used to predict whether and when bacteria will escape out of a given macrophage using observed features measured over time including intracellular and extracellular bacteria density as well as morphological features. SHAP analysis of these models will then be used to identify which features and time periods contribute to the models prediction.
Real-world systems rarely exist in isolation; they exchange energy with their environment, like a pendulum slowing from air resistance. These open, "Non-Hermitian" (NH) systems behave differently from closed ones. Two striking features are "exceptional points"—where distinct states merge—and the "NH skin effect," where states accumulate at physical boundaries. We are interested in how random "disorder" impacts NH systems. We analyse this using topology, a mathematical framework focusing on properties that survive structural deformation. Topology effectively identifies invariant characteristics in disordered systems. To verify our theories, we conduct experiments using transmission lines. These affordable, customizable circuits practically bridge electronics and photonics, which is a great advantage for studying NH systems.
The integration of perovskite solar cells (PSCs) with carbon fibre reinforced polymer (CFRP) substrates offers a route toward lightweight, high specific power structural photovoltaics for aircraft, UAVs, wind turbines, and structural batteries. However, substrate roughness and thermal mismatch challenge device fabrication. Gel-coated CFRP variants were engineered to improve surface uniformity, durability, and thermal stability, enabling stable ITO sputtering with reduced cracking. As a foundational step toward CFRP integration, triple-cation perovskites were deposited via ultrasonic spray coating combined with vacuum-assisted solution processing, yielding uniform, highly crystalline films. Incorporation of Me-4PACz enabled a simplified device architecture, delivering >15% efficiency on ITO/glass. This scalable framework establishes a pathway toward multifunctional structural energy systems.
White dwarfs (WDs) are the burned out cores of stars similar to our Sun at the end of their evolutionary path. WD binaries are binary (two) star systems where at least one component is a white dwarf. If the binary system is oriented right in our line of sight from Earth, then we are able to observe eclipses from these systems in much the same way as we do between the Sun and Moon. The time that an eclipse occurs at is a result of the period of the binary (how long the objects take to orbit each other), allowing them to act as excellent clocks, since their eclipses should arrive with regular repetition. In reality, long-term observations of these eclipses show significant variation in the time that the eclipse occurs from when is expected. The cause of this is currently unknown, however theories attribute it to two main theories, one of which being the existence of planets orbiting the system. My work aims to shed light on the origin of these eclipse time variations.
In quantum error correction (QEC), an under-researched class of errors is so-called synchronisation errors (SEs.) These include erasure errors, in which quantum information – known as qubits – are deleted at known locations, and deletion/insertion errors, where qubits are deleted/inserted randomly. Much is understood about deletion errors, but as yet little is known about insertion errors. We present an equivalence between these error types on a class of quantum codes by deriving a set of insertion conditions, and showing that these are equivalent to an already known set of deletion conditions. This addresses a longstanding open question by providing a quantum version of a nearly sixty-year-old classical problem. We also present the first instance of a protocol that can realise insertion error QEC on near-term quantum devices, which is simple and practical to implement. Together, our results address a significant gap in the theory of SEs.
Modern neutrino experiments simulate neutrino interactions using Monte Carlo-based software called event generators. In neutrino-nucleus scattering experiments, final state interactions (FSI) - where the proton involved in the initial neutrino interaction scatters further within the nucleus - can leave the detectable output vastly different from what is expected. This effect is simulated in generators using intranuclear cascade (INC) models, and accurately modelling this process is crucial for reducing systematic uncertainty in our experiments. This poster will show preliminary results of the implementation of a novel INC, called INCL++, for use in the experiment-driven event generator NEUT, showing how improvements to our FSI modelling may improve our estimations of uncertainty and widen our scope to unexplored areas of neutrino-nucleus interaction physics in the future.
Single-photon sources are crucial in the development of quantum-secure telecommunication networks. A single-photon source is a type of quantum emitter in which photons are emitted individually, one at a time. A promising class of quantum emitters is quantum dots, which are semiconductor structures formed by embedding a lower bandgap material within a higher bandgap material matrix. This structure confines charge carriers and creates discrete emission lines similar to those observed in atomic systems. This poster explores quantum dots that emit photons in the telecom C-band (1530 – 1565 nm). Emission within this wavelength range is particularly important due to minimal fibre attenuation losses in this region, making it ideal for long-distance fibre-optic communication networks.
The cell envelope is one of the most crucial components of the bacterial cell. It maintains ideal conditions within the cell, as well as determining how the bacterium interacts with its environment. This project is examining how antimicrobial compounds (β-lactam antibiotics) alter the cell wall structure of the gram-positive species Staphylococcus aureus, as well how these structural changes vary during the stationary growth phase. Recently, using small-angle neutron scattering (SANS), we have been able to investigate these structural changes in different strains of S.aureus.
Quantum dots (QDs) are semiconductor nanocrystals which can emit single photons when excited by a laser pulse, similar to individual atoms. For various applications, it is important that these single photons are indistinguishable from each other and emitted with a high efficiency. Unfortunately, QDs are susceptible to interactions with vibrations in their host crystal lattice, known as phonons, which reduce the indistinguishability of these emitted photons. Furthermore, the common approaches to mitigate these effects sacrifice efficiency. Interestingly, it has recently been proposed that these interactions can be eliminated by reaching a phonon decoupling regime where the QD emits a photon before the phonon interactions have a chance to occur. Reaching this regime would allow for high simultaneous indistinguishability and efficiency. My poster focuses on the optimization of nanoscale optical cavities using simulations to achieve phonon decoupling and explore new regimes of QD physics.
Non-covalent associations, such as the binding of small molecules to proteins, are ubiquitous in biology, and it is crucial to quantify these in order to optimise them. This can be achieved by calculating the change in enthalpy, quantifying the strength of interactions, and entropy, quantifying molecular dynamics. Entropy is more difficult to calculate as it quantifies the probability distribution of all states available to a system. This is particularly difficult for proteins as they are large, flexible molecules and the analysis of biomolecular systems requires assessing both solute and solvent terms. The aim of this work is to calculate the entropy change associated with protein-ligand binding using multiscale cell correlation. This approach partitions entropy components into independent, easier to calculate, terms and it can be applied equivalently to all molecules. A well known system, as well as two protein-protein complexes involved in pancreatic cancer, are characterized.
Bacterial chemotaxis is a mechanism for directing bacteria motility to navigate gradients of chemical concentration . In previous studies, the motility of swimming bacteria during chemotaxis, and signalling mechanisms involved have been thoroughly examined. For the first time, our group revealed that, in addition to swimming bacteria, a single surface-attached bacteria also exhibit chemotactic responses. It was shown that P. aeruginosa, performs chemotaxis by reversing its movement on surfaces. Another surface-attached bacterium whose chemotaxis has been less extensively studied is Flavobacterium johnsoniae, a member of the Bacteroidetes phylum, which moves via gliding on surfaces This psoter initially introduces the motility and sensory mechanisms of swimming and surface-attached bacteria. It then, discusses unanswered questions regarding the mechanisms underlying the chemotactic responses of F. johnsoniae to chemical gradients
Photonic crystal (PhC) nanocavities in GaAs hosting self-assembled InAs quantum dots (QDs) provide a promising platform for cavity quantum electrodynamics and scalable quantum photonic integration. Achieving independent tuning of each emitter in multi-QD systems is an important measure toward scalable quantum photonic devices. In this work, we employ a scalable electrical approach to tune spatially separated quantum dots embedded in a PhC microcavity formed in a GaAs p–i–n diode heterostructure. Targeted O⁻ ion implantation into the p-doped region creates a high-resistivity barrier that enables localized Stark tuning while preserving the optical properties of the surrounding nanophotonic structure.
Quantum dots (QDs) are semiconductor nanocrystals that act as artificial atoms, emitting light at specific energies. Because QDs grow randomly, variations in size and shape cause them to emit light at different energies. This presents a challenge for photonic quantum circuits that utilise light-mediated interactions between QDs, as multiple dots should have the same emission energy for these interactions to take place. QDs are embedded in a photonic crystal waveguide formed by etching a periodic lattice of holes which gives rise to a photonic band gap. This allows light to only travel down a central line defect. To bring multiple dots into resonance, we utilise electrical tuning. By using oxygen implantation to create resistive barriers within the p-layer of a p-i-n diode, we can electrically isolate five sections of the waveguide without changing its optical properties. This allows for independent control of individual emitters, providing a scalable architecture for quantum networks.
Second harmonic generation (SHG) is a second order nonlinear optical process whereby light is converted to double its frequency that relies on second order nonlinear susceptibility. Combining strong nonlinearity with a high refractive index allows realisation of compact devices for SHG. In this work, we utilise the 3R phase of molybdenum disulfide (MoS2), a high-refractive-index layered transition metal dichalcogenide (TMDs) from a class of so-called van der Waals crystals, to fabricate grating nanostructures to boost the SHG signal. This approach provides a promising route for enhancing optical nonlinear processes in van der Waals materials and highlights the potential of TMDs for applications in nonlinear nanophotonics.
Fluorescent Proteins (FPs) are used as probes for a variety of different microscopy techniques, yet their utility is often limited by photobleaching and phototoxicity. Recent observations have shown that co-illumination (exciting with both blue/green and near infrared (NIR) light) significantly mitigates these issues. Co-illumination has also been demonstrated to allow FPs to function as quantum sensors and as probes of molecular size via rotational diffusion experiments. In this work we employ a multimodal approach to better understand the photophysical processes which enable co-illumination to work so effectively in these fluorescent probes. Finally, we note that this characterisation technique can be applied across a diverse range of proteins, and other fluorescent probes. This offers a framework for the rational design and optimisation of next-generation emissive probes tailored for advanced microscopy and quantum applications.