143174 research outputs found
Sort by
Plasma metabolomic signatures for copy number variants and COVID-19 risk loci in Northern Finland populations
No full textCopy number variants (CNVs) are an important class of genomic variation known to be important for human physiology and diseases. Here we present genome-wide metabolomic signatures for CNVs in two Finnish cohorts—The Northern Finland Birth Cohort 1966 (NFBC 1966) and NFBC 1986. We have analysed and reported CNVs in over 9,300 individuals and characterised their dosage effect (CNV-metabolomic QTL) on 228 plasma lipoproteins and metabolites. We have reported reference (normal physiology) metabolomic signatures for up to ~ 2.6 million COVID-19 GWAS results from the National Institutes of Health (NIH) GRASP database, including for outcomes related to COVID-19 death, severity, and hospitalisation. Furthermore, by analysing two exemplar genes for COVID-19 severity namely LZTFL1 and OAS1, we have reported here two additional candidate genes for COVID-19 severity biology, (1) NFIX, a gene related to viral (adenovirus) replication and hematopoietic stem cells and (2) ACSL1, a known candidate gene for sepsis and bacterial inflammation. Based on our results and current literature we hypothesise that (1) charge imbalance across the cellular membrane between cations (Fe2+, Mg2+ etc.) and anions (e.g. ROS, hydroxide ion from cellular Fenton reactions, superoxide etc.), (2) iron trafficking within and between different cell types e.g., macrophages and (3) systemic oxidative stress response (e.g. lipid peroxidation mediated inflammation), together could be of relevance in severe COVID-19 cases. To conclude, our unique atlas of univariate and multivariate metabolomic signatures for CNVs (~ 7.2 million signatures) with deep annotations of various multi-omics data sets provide an important reference knowledge base for human metabolism and diseases
Bounding elastic photon-photon scattering at s≈1 MeV using a laser-plasma platform
No full textWe report on a direct search for elastic photon-photon scattering using x-ray and photons from a laser-plasma
based experiment. A photon beam produced by a laser wakeeld accelerator provided a broadband spectrum
extending to above = 200 MeV. These were collided with a dense x-ray eld produced by the emission from
a laser heated germanium foil at ≈ 1.4 keV, corresponding to an invariant mass of √ = 1.22 ± 0.22 MeV. In
these asymmetric collisions elastic scattering removes one x-ray and one high-energy photon and outputs two
lower energy photons. No changes in the photon spectrum were observed as a result of the collisions allowing
us to place a 95% upper bound on the cross section of 1.5 × 1015 μb. Although far from the QED prediction, this
represents the lowest upper limit obtained so far for √ ≲ 1 MeV
Bias correction of quadratic spectral estimators
No full textThe three cardinal, statistically consistent, families of non-parametric estimators to the power spectral density of a time series are lag-window, multitaper and Welch estimators. However, when estimating power spectral densities from a finite sample each can be subject to non-ignorable bias. Astfalck et al. (2024) developed a method that offers significant bias reduction for finite samples for Welch’s estimator, which this article extends to the larger family of quadratic estimators, thus offering similar theory for bias correction of lag-window and multitaper estimators as well as combinations thereof. Importantly, this theory may be used in conjunction with any and all tapers and lag-sequences designed for bias reduction, and so should be seen as an extension to valuable work in these fields, rather than a supplanting methodology. The order of computation is larger than O(n log n) typical in spectral analyses, but not insurmountable in practice. Simulation studies support the theory with comparisons across variations of quadratic estimators
Question answering with LLMs and learning from answer sets
No full textLarge language models (LLMs) excel at understanding natural language but struggle with explicit commonsense reasoning. A recent trend of research suggests that the combination of LLM with robust symbolic reasoning systems can overcome this problem on story-based question answering (Q&A) tasks. In this setting, existing approaches typically depend on human expertise to manually craft the symbolic component. We argue, however, that this component can also be automatically learned from examples. In this work, we introduce LLM2LAS, a hybrid system that effectively combines the natural language understanding capabilities of LLMs, the rule induction power of the learning from answer sets (LAS) system ILASP, and the formal reasoning strengths of answer set programming (ASP). LLMs are used to extract semantic structures from text, which ILASP then transforms into interpretable logic rules. These rules allow an ASP solver to perform precise and consistent reasoning, enabling correct answers to previously unseen questions. Empirical results outline the strengths and weaknesses of our automatic approach for learning and reasoning in a story-based Q&A benchmark
Impact of type 1 diabetes on endothelial cells derived from living donors
No full textEndothelial colony forming cells (ECFCs) derived from peripheral blood have been shown to retain disease phenotype in several conditions thus possessing great translational potential for regenerative medicine. Hyperglycaemia may alter the phenotype of ECFCs yet the characteristics of ECFCs isolated from people with type 1 diabetes (T1D) have not been described. Here, we establish whether ECFCs can be successfully isolated from donors with T1D and we characterize their functional properties. Human ECFCs were isolated from peripheral blood of up to 9 control and T1D donors. Expression of cell markers and cytokines was analyzed using immunocytochemistry, RT-PCR and ELISA. Ca2+ signaling and contraction were studied using Fluo-4-AM in cells treated with serial concentrations of histamine (1 × 10−7-1 × 10−4M). T1D ECFCs showed robust endothelial marker expression and displayed normal morphology but had a reduced size compared to those from control donors. In response to inflammatory stimuli, T1D ECFCs exhibited exaggerated pro-atherogenic/pro-inflammatory cytokine (IL-6 and MCP-1) and adhesion molecule gene induction (VCAM-1 and ICAM-1) but suppressed induction of interferon signaling markers (IP-10). Histamine stimulated a concentration-dependent increase in Ca2+ influx in ECFCs which was significantly reduced in ECFCs from T1D donors, independent of differences in H1 receptor expression levels. Histamine-induced contraction was significantly enhanced in T1D ECFCs. ECFCs from control and T1D donors exhibit distinct phenotypic differences redolent of the vascular pathologies associated with T1D. This establishes the utility of T1D ECFCs for modeling vascular complications but also highlights the need to understand the potential limitations of autologous ECFCs to treat diabetic complications
Measuring the level of aflatoxin infection in pistachio nuts by applying machine learning techniques to hyperspectral images
No full textThis paper investigates the use of machine learning techniques on hyperspectral images of pistachios to detect and classify different levels of aflatoxin contamination. Aflatoxins are toxic compounds produced by moulds, posing health risks to consumers. Current detection methods are invasive and contribute to food waste. This paper explores the feasibility of a non-invasive method using hyperspectral imaging and machine learning to classify aflatoxin levels accurately, potentially reducing waste and enhancing food safety. Hyperspectral imaging with machine learning has shown promise in food quality control. The paper evaluates models including Dimensionality Reduction with K-Means Clustering, Residual Networks (ResNets), Variational Autoencoders (VAEs), and Deep Convolutional Generative Adversarial Networks (DCGANs). Using a dataset from Leeds Beckett University with 300 hyperspectral images, covering three aflatoxin levels (160 ppn, and >300 ppn), key wavelengths were identified to indicate contamination presence. Dimensionality Reduction with K-Means achieved 84.38% accuracy, while a ResNet model using the 866.21 nm wavelength reached 96.67%. VAE and DCGAN models, though promising, were constrained by dataset size. The findings highlight the potential for machine learning-based hyperspectral imaging in pistachio quality control, and future research should focus on expanding datasets and refining models for industry application
A mixed methods evaluation of an antimicrobial prescribing clinical decision support system app
No full textThis study evaluated how the usability and accessibility of a digital antimicrobial prescribing app influences clinical decision-making. Using a convergent parallel mixed methods design, the study assessed app usage patterns with surveys and interviews to identify common barriers. Among 700 users at a tertiary hospital, 61 completed the survey (7.3% response rate), including 52 prescribers. Additionally, 20 prescribers participated in interviews. While 87% found the guidelines relevant, only 52% rated navigation as easy, and 34% reported slower decision-making compared to other clinical decision support systems (CDSS). App use peaked during morning rounds (8–11 AM). Key challenges included navigation inefficiencies (59%), technical barriers, limited onboarding, and concerns around clinical AI transparency. Interviews highlighted frustration with excessive steps and a desire for simpler guideline access. Findings highlight the need for user-friendly CDSS tools integrated into clinical workflows, and stress the importance of stakeholder co-design to improve medication safety
Safety and efficacy of bariatric or metabolic surgery in septuagenarians: a systematic review and meta-analysis
No full textBackground: Increasing life expectancy has expanded the older population of adults living with obesity. Effective and sustainable treatment strategies are essential, due to the significant impact of obesity in the elderly on healthcare systems. We aimed to determine the safety of metabolic/bariatric surgery in septuagenarians.
Methods: A comprehensive search was conducted across PubMed/MEDLINE, Embase, Cochrane Library, Google Scholar, Scopus, and Web of Science for studies published up to December 31, 2024, using Boolean logic and MeSH terms. Eligible studies involved patients aged ≥70 years undergoing primary bariatric or metabolic surgery that reported on 30-day morbidity and mortality, length of hospital stay, weight loss outcomes (BMI, %TWL, %EWL), and improvements in comorbidities. Two reviewers independently conducted study selection, data extraction, and risk of bias assessment using the ROBINS-I tool, with discrepancies resolved by a third reviewer. Publication bias was assessed via funnel plots and Egger’s test. Data were pooled using a random-effects model.
Results: Out of 5000 articles, 12 studies met the inclusion criteria, comprising 197,194 patients, including 2528 septuagenarians. Septuagenarians had a mean age of 73.9±2.6 years, and a lower preoperative BMI (42.4±8.0 kg/m²) compared to younger patients (46.0±7.8 kg/m², p<0.0001). In patients ≥70 years, the pooled 30-day morbidity was 9.0% (95% CI: 6.0–13.0%) and mortality 0.7% (95% CI: 0.0–1.2%). Compared to younger patients, those aged ≥70 years had higher odds of 30-day complications (OR 1.55, 95% CI: 1.26–1.91, p <0.0001) and mortality (OR 4.31, 95% CI: 1.53–12.11, p=0.006). No significant differences were found in hospital stay or late complications. Although postoperative BMI was similar, septuagenarians had lower %TWL and %EWL, with comparable improvements in type 2 diabetes and hypertension.
Conclusions: Our findings support the feasibility of metabolic and bariatric surgery in septuagenarians, albeit with a higher short-term risk profile, necessitating careful patient selection and management
