Access to Research at National University of Ireland, Galway
Research at National University of Ireland, GalwayNot a member yet
15712 research outputs found
Sort by
Unravelling ethanol metabolic pathways under different feeding regimes in anaerobic digestion ecosystems
No full textFeeding regimes critically influence the efficiency of anaerobic digestion, yet its impact on ethanol-type anaerobic digestion ecosystems remains inadequately understood. A comprehensive investigation into the effect of feeding regimes on ethanol-type anaerobic digestion was conducted. Sequencing batch reactor (SBR) and continuous-flow reactor (CFR) were adopted for acclimating ethanol-type anaerobic methanogenesis. During long-term operation, the CFR system achieved complete removal of ethanol and volatile fatty acids, while an accumulation of 2250.0 ± 130.0 mg COD/L acetate and 931.0 ± 184.0 mg COD/L butyrate was observed in the SBR system. The difference was probably due to the limited activity of methanogens under acid accumulation conditions in SBR. Notably, the accumulation of acetate and the reduced pH in SBR facilitated ethanol metabolism towards butyrate production by a chain-elongating bacteria: Clostridium. Additionally, acetoclastic methanogenesis was not detected in SBR, and the dominated hydrogenotrophic methanogen was Methanobacterium. Ethanol was exclusively oxidized to acetate by ethanol oxidizers, including Desulfolutivibrio and Desulfomicrobium, facilitating interspecies hydrogen transfer with Methanospirillum in CFR. The electroactive bacteria Geobacter was enriched in CFR. Moreover, the relative abundance of genes related to hydrogen transfer and direct electron transfer increased in CFR, possibly enhancing electron transfer. These results contribute to a deeper understanding of how feeding regimes shape ethanol-type anaerobic ecosystems, providing valuable insights for optimizing anaerobic digestion processes to enhance methane production and electron transfer.This study was supported by the Taighde Éireann – Research Ireland (formerly Science Foundation Ireland) and the Sustainable Energy Authority of Ireland under the SFI Frontiers for the Future Awards Programme (22/FFP-A/10346). Huanhuan Chang thanks the scholarship from the China Scholarship Council (No 202208620052). Guangxue Wu thanks the support from the Galway University Foundation.peer-reviewe
Biomarker profiling in triple-negative breast cancer
No full textTriple-negative breast cancer (TNBC) is defined as breast cancer (BC) with an absence of oestrogen receptor (ER), progesterone receptor (PR), and human epidermal growth factor receptor 2 (HER2) by immunohistochemistry (IHC). TNBC comprises up to 20% of all sporadic BCs. TNBC is a heterogeneous group of tumours with various distinct subtypes, each characterized by distinct genetic profiles, morphologies, and clinical courses. Patients with TNBC often present at a more advanced stage, with high risk of early relapse and poorer survival than non-TNBC subtypes. Attainment of a pCR is more common in TNBC and HER2-positive BC than in ER-positive BC, with the most significant association between pCR and survival observed in these aggressive subtypes. The role of immunotherapy in the treatment of early TNBC is evolving and is being evaluated in several clinical trials. Recent trials have shown that patients with TNBC may benefit from targeted immunotherapy – programmed cell death 1 (PD-1) and programmed cell death-ligand 1 (PD-L1) – especially in tumours with high levels of tumour infiltrating lymphocytes (TILs) and PD-L1 overexpression. Expression of human endogenous retrovirus K (HERV-K) envelope protein (env) has been shown to be disproportionately increased in TNBC and is associated with poor outcome.
The aims of this work are to investigate the prognostic value of several biomarkers including TILs, PD-L1 and HERV-K in TNBC, with respect to disease-free survival (DFS), metastasis-free survival (MFS), breast cancer-specific survival (BCSS) and pathological complete response (pCR) rates for patients diagnosed with TNBC at Galway University Hospitals (GUH) over a 20-year period
Hydrogen and volatile fatty acid production from macroalgae biomass and spent coffee grounds using acidogenic fermentation
No full textThis PhD thesis aimed at investigating the valorisation of both Ulva biomass and spent coffee grounds (SCG) through acidogenic fermentation. To maximize the two substrates' potential for producing H2 and volatile fatty acids, a range of pretreatment procedures were applied. The first part of this research (Chapter 3) investigated the effects of process temperature (25C, 37C and 55C), ulvan extraction and lyophilisation on production of H2 and volatile fatty acids (VFAs) from Ulva biomass. Ulvan extracted biomass performed most favourably at each temperature, achieving a H2 yield of 12.8 (± 0.2) mL H2/g VS at 25 °C, 15.0 (± 1.2) mL H2/g VS at 37 °C, and 15.3 (± 1.5) mL H2/g VS at 55 °C. Additionally, the ulvan-extracted biomass gave the highest yield of total VFAs with the peak concentrations achieved at the end of the incubation (day 4), i.e. 666.6 (± 76.0) mg HAceq/L at 25 °C, 610.7 (± 66.9) mg HAceq/L at 37 °C, and 783.7 (± 31.8) mg HAceq/L at 55 °C, thus indicating that the extraction of the polysaccharide ulvan did not limit the potential valorisation of the Ulva substrate. In chapter 4, the impact of ulvan extraction and morphotype of Ulva sp. (foliose vs tubular) on acidogenic fermentation in a fed batch bioreactor system was investigated. Regarding Ulva morphology, both morphotypes exhibited comparable VFA production when ulvan was present, with peak concentrations of VFAs recorded at 2179.5 mg HAceq/L (foliose) and 2029.3 mg HAceq/L (tubular). Moreover, the ulvan extraction had no effect on the tubular morphotype, which peaked at 2165.0 mg HAceq/L. Additionally, the hydrolysis of the tubular morphotype profited from ulvan extraction peaking, at a carbohydrate release of 9.8 g glucose/L. Chapter 5 investigated the impact various pretreatments, i.e. alkali, ultrasound and static magnetic field (SMF), have on biomolecule solubilisation and anaerobic fermentation of spent coffee grounds (SCG), with a focus on H2 and VFA production. Ultrasound pretreatment resulted in the highest accumulation of VFAs (3260.0 ± 164.2 mg HAceq/L), accompanied with an increased caproic acid concentration (1114.4 mg HAceq/L). Additionally, SMF pretreatment was found to be not a viable means of pretreatment for preparing SCG for anaerobic fermentation. Chapter 6 further evaluates the valorisation of raw and ultrasound pretreated SCG in bioreactor systems. Ultrasound pretreatment resulted in an increase of 27.4% in VFA accumulation. The microbial community analysis indicated that an elevated relative abundance of Clostridia species corresponded with the peaks in H2 production. Following fermentation, the VFA-rich effluent was used as the substrate to an upflow anaerobic sludge bed reactor (UASB) to produce CH4 (9.0 mL CH4/g VS/d) at a low pH (4 - 5)
Rapid detection of the novel human pathogen Pantoea piersonii: advancements in methodology
No full textPantoea piersonii a novel bacterium isolated from the International Space Station (ISS) presents a unique challenge for microbial monitoring in spaceflight and more recently in clinical environments. Identification of the organism currently involves culture, followed by whole genome sequencing and analysis of generated sequences. Since the MALDI-TOF profile of this pathogen is absent from the database and 16S rRNA sequencing fails to resolve its identity to the nearest neighbour, a definitive genetic marker is required for unambiguous identification of the organism. Given the increase in the number of reported clinical cases, there exists a need for a rapid method for identification of the organism which could be utilised in a range of environments including the clinical setting.
This study describes the design, development and validation of a specific and sensitive real-time PCR assay for the specific detection of P. piersonii. The assay targets a unique region of the malate dehydrogenase gene, confirmed through comparative genomic analysis. We demonstrate the performance of the assay in terms of analytical specificity, sensitivity, and robustness, ensuring its suitability for both space microbiology applications and clinical use.peer-reviewe
Mechanical and microstructural analysis of glass Fibre-Reinforced high density polyethylene thermoplastic waste composites manufactured by material extrusion 3D printing technology
No full textAs the demand for thermoplastic composite materials continues to surge across diverse industries, the imperative for efficient waste management, collection, and recycling of domestic thermoplastic materials has become increasingly evident. This study presents a novel approach to upcycling non-printable domestic High-Density Polyethylene (HDPE) waste and industrial waste fibres into viable feedstock for 3D printing manufacturing, addressing critical technical challenges such as adhesion to the build plate, nozzle clogging, and warpage. The investigation focused on the incorporation of various glass fibre weight fractions—2 wt%, 5 wt%, 8 wt%, 15 wt%, 30 wt%, 45 wt%, and 60 wt%—into HDPE waste. The optimisation of filament production and 3D printing parameters was performed, resulting in the highest tensile strengths for both filament (54.5 MPa) and printed specimens (29.83 MPa) at a glass fibre content of 45 wt%. It was observed that exceeding a glass fibre content of 45 wt% led to fibre agglomeration, subsequently diminishing tensile strength in both filaments and printed samples. In contrast, the samples with 30 wt% glass fibre exhibited the highest flexural strength. The study further employed optical microscopy to evaluate fibre distribution and internal defects within the filaments and printed samples, while Scanning Electron Microscopy (SEM) was utilised to analyse the fracture surfaces of 3D printed coupons. The findings revealed that the dominant fracture mechanism in these composites was fibre pull-out. The research successfully fabricated a demonstrator, showcasing the potential of HDPE domestic thermoplastic waste, reinforced with glass fibre, to be processed into functional industrial composite components via 3D printing.This project has received funding from the European Union’s Horizon 2020 research and innovation programme under the Marie Skłodowska-Curie grant agreement No. 847402. This publication has also emanated in part from research conducted with the financial support of Taighde Éireann – Research Ireland, formerly Science Foundation Ireland under Grant number 22/IRDIFB/10946, 16/RC/3872 and 21/RC/10295_P2. This research was also supported by the Department of Business, Enterprise and Innovation and administered by Enterprise Ireland under the Disruptive Technologies Innovation Fund, MI-DRONE Project (Contract Ref: DT 2020 0221). For the purpose of Open Access, the author has applied a CC BY public copyright licence to any Author Accepted Manuscript version arising from this submission.peer-reviewe
Development of a shoreline change modelling system for Brandon Bay, Ireland using tide, wave and morphology models
No full textNumerical models can play an important role in better understanding the coastal erosion process and the relative impacts of its drivers such as winds, waves and tides. The primary aim of this research was to develop a modelling system for simulation of shoreline change in Brandon Bay, located in Co. Kerry on the western coast of Ireland, and use the model to gain a deeper understanding of current circulation and wave climate in the bay and their impact on shoreline change. A secondary aim was to use the modelling system to study the potential impacts of future climate change on shoreline erosion/acretion. Hydrodynamic, spectral wave, and shoreline change models were developed using the well-known software packages Delft3D, SWAN and XBeach, respectively, and were validated against field measurements. The models revealed a strong wind influence in Brandon Bay, especially in its eastern half. This persists all year but is strongest during winter and is present even in the bottom layers. The model results showed that mean current speeds when wind was included in the model were 3-5 times higher than those when wind was omitted. The wave climate was characterised using a 7-year modelled wave dataset. Results showed significant seasonal variability in significant wave heights; winter means in the east of the bay ranged from 2-3 m compared to 0.5-1.0 m for summer. Five different storm wave classification approaches were used to identify storm wave characteristics from the modelled wave data. A fixed threshold approach, was found to be more effective for short-term periods (1-3 years), while a statistical-based classification using exceedance values was more suitable for longer periods (3+ years). The choice of approach had a significant effect on the number of identified storm waves which ranged from 36 to 136 for the most appropriate approaches. Coastline change was examined using cross-sectional profile datasets from surveys carried out between October 2021 and November 2022, paired with numerical model simulations. Erosion occurs mostly in the winter months, followed by natural recovery in the milder spring and summer seasons. The future climate model simulation showed that more intense storms in the future with increased wave height could cause increased coastal erosion and could the occurrence of more frequent storm waves
MuRelSGG: Multimodal relationship prediction for neurosymbolic scene graph generation
No full textNeurosymbolic Scene Graph Generation (SGG) is a promising approach that jointly leverages the perception capabilities of deep neural networks and the reasoning capabilities of symbolic techniques for scene understanding and visual reasoning. SGG systematically captures semantic components, including objects and their relationships, in images, enabling structured representations of visual data. However, existing SGG methods exhibit constrained accuracy and limited expressiveness, particularly in long-tail relationship prediction. To address these limitations, we present MuRelSGG, a novel neurosymbolic SGG framework that integrates a Transformer-based multimodal relationship prediction pipeline with common sense knowledge enrichment. This synergistic combination encapsulates global context, long-range dependencies, and complex object interactions to enhance relationship prediction in SGG. The proposed neurosymbolic architecture begins with object detection via Faster R-CNN, followed by a cascade of Multi-Head Attention Transformers (M-HAT) and Vision Transformers (ViT) for relationship prediction. Subsequently, CSKG enrichment refines and augments visual relationships, improving both accuracy and expressiveness. We conduct extensive evaluations on both the Visual Genome (VG) and GQA datasets to assess performance and generalizability. MuRelSGG achieves substantial gains in recall rates (VG: R@ 100=43.2 , mR@ 100=14.9 ; GQA: R@ 100=42.1 ), outperforming state-of-the-art SGG techniques. Ablation studies confirm the critical contributions of M-HAT, ViT, linguistic features, CSKG enrichment and embedding similarity thresholds, demonstrating the effectiveness of structured knowledge integration for long-tail relationship prediction. These findings underscore the potential of combining deep learning architectures with structured knowledge bases to advance visual scene representation and reasoning.peer-reviewe
Integrated multi-index drought monitoring and projection under climate change
No full textUnderstanding drought trends under climate change is critical for effective water resources management. Given the complex nature of droughts, relying solely on a single variable for drought analysis might not be adequate for promptly and reliably detecting drought conditions. This study introduces a comprehensive approach to drought monitoring and projection under climate change. The methodology assesses drought conditions by considering different key factors such as soil moisture, precipitation, runoff, relative humidity and (unmet) water demands. By considering these variables, the study aims to provide a more holistic understanding of drought dynamics. To achieve this, a range of indices are used including the Standardised Precipitation Index (SPI), the Standardised Soil Moisture Index (SSI), the Standardised Relative Humidity Index (SRHI), the Standardised Runoff Index (SRI), the Multivariate Standardised Drought Index (MSDI), the Inflow-Demand Reliability Indicator (IDRI), the Water Storage Resilience Indicator (WSRI) and the Multivariate Standardised Reliability and Resilience Index (MSRRI). Furthermore, the study employs a (non-)parametric set of copula functions to analyze compound drought events, which consider the interconnected nature of different drought characteristics. These indicators detect drought onset, persistence, and spatial extent over Europe for different future climate scenarios, using data from the Coupled Model Intercomparison Project Phase 6 (CMIP6). The study explores future climatic changes and their potential impact on drought patterns, comparing them with historical data. Results indicate that drought estimates vary significantly across different locations, time periods, and types of drought. The study captures this complexity and provides a more nuanced and resilient framework for understanding drought risks across different contexts. For example, results show that in Dublin, Ireland, the used climate scenarios project general wet meteorological conditions until 2050 (SPI > 0). However, extreme hydrological droughts are anticipated during the time based on SRI (SRI = -1.8). In addition, considering MSRRI, more socio-economic droughts are expected for Dublin by 2050. This study serves as a valuable resource for hydrologists, policymakers, and risk managers, offering guidance on understanding drought dynamics and informing decision-making processes related to drought prevention and mitigation strategies.This study was funded by the Irish Research Council (IRC), Ireland (project code: GOIPD/2023/1627).peer-reviewe
Microstructural buckling in soft visco-hyperelastic laminates
No full textIn this work, we study the role of visco-(hyper)elasticity in the microstructural buckling of soft laminates under compressive loads. We find that the onset of buckling is related to the contrast in specific stress components; this allows us to develop analytical estimates for the critical loading. Our numerical analysis provides details on the dependence of critical strain and wavelength on the loading strain rate. We show that by activating the viscoelasticity of the stiffer layer with an increasing strain rate, one can promote the early development of the buckling (the critical strain decreases as the strain rate is increased). The tunability of the critical strain is bounded by the limits for fast and slow loading rates. Furthermore, the buckling wavelength can be tuned through strain rate variability; this effect is stronger in laminates with lower volume fractions (of stiff layer phase), while for high volume fractions, laminates tend to develop longer wavelength instabilities with diminishing tunability.We gratefully acknowledge the support of the European Research Council (ERC) under Grant No. 852281-MAGIC.peer-reviewe
Synthetic face ageing: Evaluation, analysis and facilitation of age-robust facial recognition algorithms
No full textEstablishing the identity of an individual from their facial data is widely adopted across the consumer sector, driven by the use of facial authentication on handheld devices. This widespread use of facial authentication technology has raised other issues, in particular those of biases in the underlying algorithms. Initial studies focused on ethnic or gender biases, but another area is that of age-related biases. This research work focuses on the challenge of face recognition over decades-long time intervals and explores the feasibility of utilizing synthetic ageing data to improve the robustness of face recognition models in recognizing people across these longer time intervals. To achieve this, we first design a set of experiments to evaluate state-of-the-art synthetic ageing methods. In the next stage, we explore the effect of age intervals on a reference face recognition algorithm using both synthetic and real ageing data to perform rigorous validation. We then use these synthetic age data as an augmentation method to facilitate the age-invariant face recognition algorithm. Extensive experimental results demonstrate a notable improvement in the recognition rate of the model trained on synthetic ageing images, with an increase of 3.33% compared to the baseline model when tested on images with a 40-year age gap. Additionally, our models exhibit competitive performance when validated on benchmark cross-age datasets and general face recognition datasets. These findings underscore the potential of synthetic age data to enhance the performance of age-invariant face recognition systems.Irish Research Council (Grant Number: EPSPG/2020/40 and IRCLA/2023/1992)peer-reviewe