23343 research outputs found
Sort by
Matrix Metalloproteinases, Tissue Inhibitors of Metalloproteinases, and Their Ratios in Women with Polycystic Ovary Syndrome and Healthy Controls
Get PDFMatrix metalloproteinases (MMPs) are M2 macrophage markers that are modulated by inflammation. A disintegrin and metalloproteinases (ADAMS) and those with thrombospondin motifs (ADAMTS) regulate the shedding of membrane-bound proteins, growth factors, cytokines, ligands, and receptors; MMPs, ADAMS, and ADAMTS may be regulated by tissue inhibitors of metalloproteinases (TIMPs). This study aimed to determine whether these interacting proteins were dysregulated in PCOS. A Somascan proteomic analysis of 12 MMPs, three of their inhibitors (TIMP-1, 2, 3), two ADAMS (9, 12), five ADAMTS (1, 4, 5, 13, 15), insulin-like growth factor binding protein-1 (IGFBP-1), and insulin-like growth factor-1 (IGF-1) was undertaken in a well-validated PCOS database of 143 women with PCOS and 97 controls. Women with PCOS had significantly higher levels of MMP-9 and lower levels of MMP-2, MMP-14, TIMP-2, IGFBP-1, and IGF-1 compared to the controls (p < 0.0001, p < 0.005, p < 0.04, p < 0.05, p < 0.0001, and p < 0.0001, respectively). No differences were observed for any other MMPs. The ADAMS or ADAMTS levels did not differ between groups. Body mass index (BMI) was correlated with MMP-9 (p < 0.01), MMP-1 (p < 0.05), MMP-2 (p < 0.05), MMP-10 (p < 0.005), MMP-12 (p < 0.005), ADAM-9 (p < 0.05), and IGFBP-1 (p < 0.0001), but only MMP-9 still differed after accounting for BMI. MMP-9/TIMP-1, MMP-9/TIMP-2, and MMP-9/TIMP-3 ratios were higher in the PCOS group (p < 0.01), whilst MMP-17/TIMP-1 and MMP-17/TIMP-2 were lower (p = 0.01). MMP-2/TIMP ratios showed no difference between groups. TIMP-2 was positively correlated with CRP (p < 0.01). MMP changes in PCOS are largely driven by BMI, though increased MMP-9 is BMI-independent, suggesting that any deleterious effects of MMP-9 would be potentially exacerbated by a concomitantly increased BMI. The significant increases in the MMP-9/TIMP ratios suggests MMP-9 overactivity in PCOS
Decellularised natural biomaterials for reconstructive bioengineering of the lower urinary tract
Get PDFSurgical reconstruction is required for congenital and acquired diseases affecting the lower urinary tract in adult and paediatric patients. In conditions such as small-contracted urinary bladders linked to infections, incontinence and renal damage, and hypospadias where baby boys’ external urethral orifice is abnormally positioned, the lack of tissue availability and quality can compromise reconstructive outcomes, making the case for biomaterials with tissue-integrative properties.Natural biomaterials comprise extracellular matrix proteins that are conserved among different species. Such biological tissues possess the strength of the tissues from which they have been derived and can serve as scaffolds for tissue regeneration, providing a natural microenvironment for cell attachment, migration, and proliferation.A non-cross-linked porcine acellular matrix (PABM) has been previously developed by the decellularisation of full-thickness porcine bladders through a series of hypotonic buffers, detergents and enzymes to remove soluble cell and DNA content. The biomatrix retains the major structural components and physical characteristics, in terms of compliance and strength, of the originating organ tissue, and resulted biocompatible and with potential for use in urological surgery and tissue-engineering applications.The need for a scalable manufacturing process that allows for the transfer of the porcine biomatrix production from the benchtop to an industrial scale was identified.The overall aim of the present research was to investigate the PABM's biomaterial integration properties and tissue regeneration characteristics when applied in homologous surgical settings. It was hypothesised that this acellular biomatrix would promote an adequate initial mechanical support and favourite a biological environment able to promote tissue regeneration as an implant in in vivo models.PABM was therefore prepared testing different decellularisation techniques for full-thickness porcine urinary bladders and the suitability for scaling up the decellularisation process was investigated.Biological response parameters for this non-cross-linked natural material were determined and compared to a commercially available cross-linked biomatrix following in vivo surgical application.PABM integration properties and tissue regeneration characteristics were examined in a homologous surgical setting where the PABM was implanted in a large animal model of urinary bladder reconstruction.This study contributed to demonstrate the feasibility of a commercial manufacturing production for decellularised full-thickness porcine bladder tissue, and it produced evidence that PABM provides an anti-inflammatory, pro-regenerative tissue-integrative environment in in vivo settings. Based on its specific characteristics, PABM was here shown to be a candidate biomaterial for surgical applications in urinary tract reconstruction and beyond, with potential to contribute to ameliorating quality of life in paediatric patients worldwide
Addressing Explainability of Generative AI using SMILE
Get PDFThe rapid advancement of generative artificial intelligence has enabled models capable of producing complex textual and visual outputs; however, their decision-making processes remain largely opaque, limiting trust and accountability in high-stakes applications. This thesis introduces gSMILE, a unified framework for the explainability of generative models, extending the Statistical Model-agnostic Interpretability with Local Explanations (SMILE) method to generative settings. gSMILE employs controlled perturbations of textual input, Wasserstein distance metrics, and weighted surrogate modelling to quantify and visualise how specific components of a prompt or instruction influence model outputs. Applied to Large Language Models (LLMs), gSMILE provides fine-grained token-level attribution and generates intuitive heatmaps that highlight influential tokens and reasoning pathways. In instruction-based image editing models, the exact text-perturbation mechanism is employed, allowing for the analysis of how modifications to an editing instruction impact the resulting image. Combined with a scenario-based evaluation strategy grounded in the Operational Design Domain (ODD) framework, gSMILE allows systematic assessment of model behaviour across diverse semantic and environmental conditions. To evaluate explanation quality, we define rigorous attribution metrics, including stability, fidelity, accuracy, consistency, and faithfulness, and apply them across multiple generative architectures. Extensive experiments demonstrate that gSMILE produces robust, human-aligned attributions and generalises effectively across state-of-the-art generative models. These findings highlight the potential of gSMILE to advance transparent, reliable, and responsible deployment of generative AI technologies
The genome sequence of the rudd, Scardinius erythrophthalmus (Linnaeus, 1758) (Cypriniformes: Leuciscidae)
No full textWe present a genome assembly from an individual Scardinius erythrophthalmus (Rudd; Chordata; Actinopteri; Cypriniformes; Leuciscidae). The assembly contains two haplotypes with total lengths of 1 229.65 megabases and 1 189.97 megabases. Most of haplotype 1 (95.02%) is scaffolded into 25 chromosomal pseudomolecules. Haplotype 2 was assembled to scaffold level. The mitochondrial genome has also been assembled, with a length of 16.61 kilobases. This assembly was generated as part of the Darwin Tree of Life project, which produces reference genomes for eukaryotic species found in Britain and Ireland
Refining Reconstruction: Discovering the capabilities and limitations of pollen analysis using the Multiple Scenario Approach
No full textThis thesis contains several levels of exploration of pollen data quality, in particular related to the Multiple Scenario Approach (MSA), a method for spatially reconstructing landcover from pollen data, and the use of confidence intervals. The primary objective of this project has been to develop new software for applying the MSA, while the aim has been to open the “black box” of pollen data, spatial pollen modelling, and the MSA in order to improve interpretation and prevent misuse.First, the history and assumptions leading up to the MSA are explored. Then, the justification for requiring new software and what changes were made in comparison to the existing software (HUMPOL_0) are set out, followed by several tests that determine whether MSA-Q works as intended. Then, a novel method for determining pollen concentration confidence intervals using Monte Carlo iterations is introduced. This is followed by a new method for determining Cut-off Values for the metric currently used to determine when a MSA scenario should be accepted or not: Square Chord Distance. Finally, all knowledge and novel methods introduced in this thesis are applied to a case study of two existing pollen records that span the Neolithic and Bronze Age, from Sutton Common, UK, in which the specifics investigated are whether the two cores are sensitive to clearance and anthropogenic activity in the upland Magnesian Limestones, and how the wetlands around the site have developed over time.It is concluded that the MSA and the new software MSA-Q can be used to investigate spatially explicit hypotheses that follow from traditional pollen analysis, and are especially useful for investigating and displaying equifinality. The new method for determining Square Chord Distance Cut-off Values for the MSA further improves on this by better aligning the model outputs with pollen data quality and pollen assemblage features, and a new method for displaying MSA output improves the way in which equifinality can be visualized. The Open Science, FAIR approach to the development of MSA-Q ensures that the MSA is available to those who wish to use it and that the method can be scrutinized and improved by anyone who wishes to do so. It also ensures that MSA-Q can be applied using a wide selection of systems, including those for High-Performance Computing, resulting in a significantly improved workflow. MSA-Q is available on GitHub: https://github.com/ThyavdBerg/MSA-QIt is also concluded that calculating and visualizing confidence intervals for pollen data can improve visual interpretation of traditional pollen stratigraphic diagrams and may tentatively be especially helpful for early career scientists. The new Monte Carlo method for calculating pollen concentration confidence intervals improves upon this by providing a method that can be used for low total pollen counts and low individual pollen type counts, without requiring counting techniques that are different from current established methods.Finally, from the Sutton Common case study, it is concluded that the existing pollen records from Hampole Beck and Shirley Wood do not capture any clearance and anthropogenic activity from the Magnesian Limestones, and only capture spatial details from the local wetlands and general information from the surrounding woodlands. It is found that the wetland at the site became progressively more open from the Neolithic through to the Bronze Age, likely in relationship with Relative Sea Level rise, but also that despite this increased openness the local site would likely not have been traversable on any other way than by foot, and was unlikely to have been used for livestock, due to presence of dense Alnus carr
Vascular, inflammatory and perceptual responses to hot water immersion: Impacts of water depth and temperature in young healthy adults
No full textRepeated hot water immersion can improve cardiovascular health; however, the respective effects of distinct immersion protocols remain unclear. Twenty-two healthy adults completed three 30-min hot water immersion bouts of different water temperatures and immersion depths (40°C shoulder-deep immersion, 40-Shoulder; 42°C waist-deep immersion, 42-Waist; and 40°C waist-deep immersion, 40-Waist) in a randomised crossover design. Vascular, inflammatory and perceptual responses were collected via brachial and superficial femoral artery ultrasound, venous blood sampling and perceptual scales. Rectal temperature increased less in the 40-Waist (Δ0.5 ± 0.1°C) condition than the other conditions (40-Shoulder: Δ0.9 ± 0.3°C, 42-Waist: Δ0.9 ± 0.3°C, P < 0.001). Arm skin temperature increased more in the 40-Shoulder (Δ5.2 ± 1.9°C) condition than the other conditions (40-Waist: Δ2.6 ± 1.0°C, 42-Waist: Δ3.6 ± 1.1°C, P < 0.001), whilst thigh temperature had a greater increase in the 42-Waist (8.6 ± 1.3°C) condition than either the 40-Waist (7.8 ± 0.2°C) or 40-Shoulder (Δ7.8 ± 1.0°C) conditions (P < 0.001). Brachial artery shear rate was greatest post-immersion following the 40-Shoulder condition (40-Shoulder: Δ121 ± 94/s, 42-Waist: Δ47 ± 73/s, 40-Waist: Δ−21 ± 41/s, P < 0.001) whereas superficial femoral artery shear rate was largest following the 42-Waist condition (40-Shoulder: Δ143 ± 61/s, 42-Waist: 196 ± 85/s, 40-Waist: 131 ± 93/s, P < 0.001). IL-6 (P = 0.16) and cortisol (P = 0.83) responses did not differ between conditions. Perceptual responses were more favourable in the 40-Waist condition. Taken together, these data demonstrate that the distinct region-specific arterial responses align with increases in local skin temperature to alterations in hot water immersion protocols, whilst showing that beneficial physiological responses may be accompanied with less favourable perceptual responses
Environmental ecomusicology in class: Empirical evidence and reflections on sensitizing music(ology) students to climate change
No full textA Longitudinal Ordinal Collaboration Network Model for Multiclass Link Prediction and Its Application
No full textWith the growing application of collaboration networks in scientific research, link prediction has become an essential tool for understanding the structural and evolutionary mechanisms of academic relationships. Traditional link prediction methods typically model links as a binary classification problem, predicting whether a link exists between a pair of nodes. However, in real-world networks, collaboration intensity varies, and simple 0–1 classification fails to capture such heterogeneity. To address this limitation, we propose a longitudinal ordinal collaboration network model for multiclass link prediction, which accounts for both the ordinal nature of collaboration intensity and the temporal evolution of collaborative relationships. Under the maximum likelihood framework, an (Formula presented.) regularization term is incorporated to enhance the stability and interpretability of parameter estimation. The proposed model extends link prediction from binary to ordinal multiclass classification and explicitly introduces a longitudinal structure to characterize the evolution of collaborations over time. Simulation studies and empirical analyses demonstrate that the model offers strong explanatory power and predictive performance, providing an effective tool for dynamic modelling of collaboration networks
An options-pricing approach to forecasting the US election
Get PDFA subjective probability argument suggests vote-share estimates from polls can be interpreted as market prices. The corresponding election constitutes the price at a known future date. This makes an options-pricing approach particularly attractive. The approach works well for proportional systems. Here, we show how to adjust the approach for non-proportional first-past-the-post systems. We illistrate our approach with an application to the most recent 2020 and 2024 elections