College of Science, Engineering and Food Science - Masters by Research Theses
Permanent URI for this collection
Browse
Recent Submissions
Item Simulating the performance of SAW-based gas sensors(University College Cork, 2026-03-31) Acharya, Suman; Nagaraja, Veda Sandeep; O'Riordan, Alan; Pillai, Gayathri; Research IrelandSurface Acoustic Wave (SAW) devices offer a promising route for compact, sensitive, and low-power gas sensors suitable for real-time greenhouse gas monitoring. This thesis investigates the design and optimization of piezoelectric SAW resonators for the detection of methane (CH₄) and nitrous oxide (N₂O), with particular focus on lithium niobate (LiNbO3) and lithium tantalate (LiTaO₃) substrates. The work first examines the acoustic behavior of 128°-YX LiNbO₃ resonators, a material orientation known for strong electromechanical coupling and large surface displacement. 3D Finite Element Analysis (FEA) were carried out in COMSOL Multiphysics to study the influence of interdigitated transducers (IDTs) geometry, aperture, delay-line length, metallization ratio, and electrode configuration across different operating frequencies. The results show that 128°-YX LiNbO₃ can support closely spaced Rayleigh and shear-horizontal modes, with their relative dominance depending strongly on device frequency and geometry. At lower frequencies, careful optimization of the IDT design suppresses unwanted modes and promotes Rayleigh-wave operation, which is favorable for surface-sensitive gas sensing. At higher frequency, however, the shear-horizontal mode remains dominant, indicating that 128°-YX LiNbO₃ requires careful design consideration for high-frequency gas sensing applications. The thesis also investigates spurious-mode suppression in 42°-YX LiTaO₃ devices using tilted IDT configurations. Both simulation and experimental results show that electrode tilt can reduce unwanted modes, although this improvement is accompanied by a reduction in electromechanical coupling. A modified Butterworth–Van Dyke equivalent circuit model was further developed to extract device parameters from measured admittance responses and support comparison between simulation and experiment. Finally, optimized SAW structures were evaluated for CH₄ and N₂O sensing using metal–organic frameworks (MOFs) sensing layers, including Zeolitic Imidazolate Framework-8 (ZIF-8) and Zirconium based MOF (UiO-66). Frequency-domain simulations confirm that gas adsorption produces measurable resonance shifts through mass-loading effects, with ZIF-8 showing stronger sensitivity than UiO-66 under the investigated conditions. Overall, this work provides a systematic understanding of material orientation, acoustic mode behavior, electrode design, and sensing-layer effects in LiNbO₃- and LiTaO₃-based SAW devices, supporting their further development for compact greenhouse gas sensing platforms.Item Biodiversity in urban landscapes – assessing trimmed habitats and wild zones within the UCC campus(University College Cork, 2025-09-21) O'Mahony, Paula; Henriques, Rossana; Butler, Fidelma; McKeown, Michelle; University College CorkAlthough urban landscapes present significant challenges for pollinator biodiversity, they can also support diverse plant–pollinator communities when managed properly. To address this question, I investigated how four different grassland management regimes implemented at the University College Cork campus (Ireland) affected pollinator activity, floral diversity, pollen viability, and soil organic matter (SOM) content. From June to September 2024, fieldwork was carried out weekly across Highly Managed (HM), No Mow (NM), Planted Meadow (PM), and Unmanaged (UM) sites. Using a wide range of methods (e.g., Flower-Insect-Timed (FIT) counts, floral surveys, pollen viability staining), I observed a total of 737 pollinator visits, with activity varying significantly between management regimes (χ² = 40.73, p < 0.001). Highly Managed sites sustained the lowest diversity and visitation, while Planted Meadows and Unmanaged sites supported the highest. Seasonal activity peaked in mid-summer, and pollen analysis confirmed high viability (>80%) across most wildflower species, underscoring their value as nutritional resources. SOM was highest in Unmanaged sites and lowest in Planted Meadows, reflecting long-term differences in management and organic inputs. The findings of this study suggest that targeted, low-intervention practices (e.g., extending no-mow periods, creating planted meadows, and retaining areas of natural succession) contribute to increased pollinator visitation, enhanced floral resources, and support for viable pollen production in urban grasslands. By demonstrating a clear link between floral diversity, pollinator activity, and resource quality, the results show how management choices directly shape ecological outcomes. Together, these findings provide evidence-based guidance for urban biodiversity strategies, complementing initiatives such as the All-Ireland Pollinator Plan, and illustrate the potential for universities and other institutions to function as living laboratories for conservation efforts.Item Combined CDK12/13 and BET inhibition as a therapeutic strategy in breast cancer(University College Cork, 2026-03-31) Rehman, Faisal; Krajewska, Malgorzata; O'Connor, RosemaryBackground: Breast cancer is the most common type of malignancy among women. Triple Negative Breast cancer (TNBC) is an aggressive subtype of breast cancer characterised by high genomic instability and limited targeted therapeutic options. Transcription Cyclin-dependent kinases, CDK12/13, are serine/threonine kinases that regulate the RNA Polymerase II elongation, mRNA processing and expression of DNA damage response genes. Dysregulation of CDK12/13 contributes to tumour progression and genomic instability. A study from our lab indicated potential for CDK12/13 inhibition as a therapeutic approach for TNBC. The aim of this thesis is to investigate the effect of dual inhibition of CDK12/13 with the JQ1 inhibitor of Bromodomain and extra-terminal domain (BET) family member BRD4 in TNBC model. Methods: Publicly available transcriptomic and proteomic clinical datasets were used to analyse the CDK12 and CDK13 expression patterns in breast cancer. Molecular modelling using AutoDock Vina was performed to investigate drug selectivity. Cell viability was evaluated using MTT and Crystal violet assays. Protein expression was analysed using western blotting. Apoptosis was analysed using Annexin V and Propidium iodide (PI) staining, while cell cycle distribution was analysed by propidium iodide (PI) staining using flow cytometry. Results: Analysis of transcriptomic and proteomic clinical data revealed that CDK12 and CDK13 are significantly overexpressed in breast cancer tissue compared to normal tissue. Molecular modelling studies demonstrated that SR4835 binds within the ATP-conserved binding pocket of CDK12/13. The sequence alignment analysis revealed 96.9% similarity between CDK12/13 in their protein kinase domain. Combination of SR4835 and JQ1 significantly reduced cell viability, induced DNA damage, promote caspase-dependent apoptosis, and arrested the cell cycle more than single agent treatment. Notably, SR4835 alone induced STAT3 phosphorylation and co-treatment with SR4835 and a neutralising IL6 antibody significantly suppressed STAT3 phosphorylation. Finally, SR4835 alone pronouncedly increased pSTAT3 phosphorylation, while JQ1 reduced it, and co-treatment reversed SR4835-induced STAT3 phosphorylation. Significance: Our study demonstrated that dual inhibition of CDK12/13 and BRD4 significantly reduced TNBC proliferation by inducing cell death and apoptosis, and blocked activation of the compensatory STAT3 pathway induced by CDK12/13 inhibition. This dual transcription-targeted strategy provides a promising therapeutic approach for TNBC and holds potential for further in vivo studies.Item An empirical study of the effect of variable network performance on teleoperated driving(University College Cork, 2026-03-31) Javed, Farhan; O'Driscoll, Aisling; Sreenan, Cormac; Hines, Andrew; Science Foundation IrelandThis thesis introduces an empirical study focused on the impact of variable network performance, specifically latency and packet loss, on the Quality of Experience (QoE) of remote drivers in Teleoperated Driving (ToD). In recent years, the development of 5G networks and autonomous driving systems has made significant strides. Yet, ToD remains a crucial backup for ensuring vehicle safety in complex driving scenarios that autonomous systems cannot handle. This study investigates the thresholds of network conditions beyond which ToD becomes unsafe, providing valuable insights into the relationship between network Quality of Service (QoS) and the remote driver’s experience. Utilising the TELECARLA simulator, a network-aware extension of the CARLA simulator, this thesis emulates network impairments and analyses their impact on driver performance. Key performance metrics, including Time to Collision (TTC), Steering Wheel Reversal Rate (SWRR), and subjective ratings like the Mean Opinion Score (MOS) and NASA-TLX workload assessment, are used to assess how varying network conditions influence the remote driver’s control and cognitive workload. Results reveal critical thresholds for network delays and packet loss, identifying the point at which remote driving becomes unsafe for drivers both with and without prior virtual driving experience. This research provides essential empirical data on the coupling of QoS and QoE in ToD systems, contributing to developing safe and reliable ToD frameworks. Furthermore, it offers practical guidance on how prior experience in virtual driving can help mitigate the effects of network variability, improving driver performance and safety in remote operations.Item Optimising transfection methods for genome editing using iPSCs and CRISPR/Cas9(University College Cork, 2026-02-28) Beltran, Jhoan Sebastian; Matsa, Elena; Lee, CiaranThe emergence of induced pluripotent stem cells (iPSCs), together with major advances in genome editing technologies, has significantly expanded the potential of cell-based therapies in regenerative medicine and immunotherapy. Since their first description by Shinya Yamanaka and Kazutoshi Takahashi in 2006, iPSCs have been widely explored as a promising platform for the treatment of a broad range of diseases, including Parkinson’s disease, Alzheimer’s disease, and organ failure. Their capacity for unlimited self-renewal and differentiation into virtually any somatic cell type make them an attractive and versatile cell source for therapeutic applications. Despite these advances, significant clinical challenges remain. One of the principal barriers to the clinical translation of iPSC-derived therapies is the potential immune rejection associated with allogeneic cell products. Consequently, multiple strategies have been proposed to mitigate immune recognition and improve the compatibility of transplanted cells. This project aimed to contribute towards the development of hypoimmune iPSC lines by targeted knock-out of human leukocyte antigen (HLA) class I and class II molecules. To achieve this, the study evaluated alternative approaches for the delivery of genome editing cargo into iPSCs, assessing their potential as less invasive alternatives to electroporation. Transfection efficiencies obtained through cell-penetrating peptide (CPP)-mediated delivery and Lipofectamine-based methods reached approximately 40–50%%, with CPPs allowing higher viability and preservation of cellular morphology. In contrast, photoporation yielded lower transfection efficiencies, averaging around 22%. Overall, these findings suggest that CPPs may provide improved balance between cargo uptake and viability for iPSCs and will support future efforts to generate hypoimmune cell lines research and therapeutic applications.