Oxford University Research Archive

University of Oxford

Oxford University Research Archive
Not a member yet
    324139 research outputs found

    Evolution of the tuberculin skin test reveals generalisable Mtb -reactive T cell metaclones

    Get PDF
    T cells contribute to immune protection and pathogenesis in tuberculosis, but measurements of polyclonal responses have failed to resolve correlates of outcome. We report the temporal evaluation of the human in vivo clonal repertoire of Mycobacterium tuberculosis (Mtb)-reactive T cell responses, by T cell receptor (TCR) sequencing at the site of the tuberculin skin test, as a model for a standardised antigenic challenge. Initial non-selective recruitment of T cells is followed by enrichment of Mtb-reactive clones arising from oligoclonal T cell proliferation. We introduce a modular computational pipeline, Metaclonotypist, to sensitively cluster distinct TCRs with shared epitope specificity, which we apply here to establish a catalogue of public Mtb-reactive HLA-restricted T cell metaclones. Although most in vivo Mtb-reactive T cells are private, 10 metaclones were sufficient to identify Mtb-T cell reactivity across our study population (N≥128), indicating striking population level immunodominance of specific TCR-peptide interactions that may inform patient stratification and vaccine development

    Computational approaches to antibody library design and property prediction

    Get PDF
    Over the past few decades, antibodies have established themselves as powerful therapeutics, used to treat viral infections, autoimmune diseases, and many cancers. However, designing and optimising such antibody drugs is still costly and time-consuming. In this thesis, we aim to demonstrate how computational methods- both old and new- can improve this process. First, we introduce Paragraph, our graph-based paratope prediction tool that offers accurate residue-level predictions in just a tenth of a second. Paragraph can help guide computational docking experiments or focus researchers’ attention on where optimising mutations should be made. Next, we explore how large language models and other tools can design antibody libraries enriched in high-affinity variants from just a single known binding sequence. We also demonstrate that once a few hundred of these variants have been tested experimentally, we can train simple machine-learning methods on this data to help screen future variants and further enrich our antibody library. Finally, we introduce Humatch, our humanness classifier and humanisation tool. Humatch uses a combination of three highly accurate CNNs and germline data to offer rapid experimental-like humanisation and ensure final heavy-light designs remain well-matched. All tools and methods presented in this thesis are easily accessible to other researchers. We hope that others may use and build upon these developments to continue to improve antibody therapeutic development

    Statistical learning of semantic and graphotactic regularities: evidence from artificial orthography learning experiments

    Get PDF
    Written languages are complex. English orthography not only reflects phoneme-grapheme mappings, but it also includes other regularities related to form-meaning mappings and graphotactics. Pseudoword experiments have shown that children and adults are sensitive to these patterns, which can influence their spelling choices even when they have not been explicitly taught the patterns or are unable to verbalise them. From this perspective, spelling acquisition can be viewed as a process of statistical learning, in which individuals become increasingly sensitive to regularities and quasi-regularities in their writing system as they gain more experience with written language. Using the artificial orthography learning paradigm, the overarching aim of this thesis is to examine whether adult spellers can simultaneously learn semantic regularities (i.e., spelling patterns that carry meaningful information) and graphotactic regularities (i.e., spelling patterns concerning permissible grapheme combinations) in new writing systems through exposure, and how this evidence informs our understanding of statistical learning in reading and spelling acquisition. Three studies were conducted to address these questions. Study 1 presents a systematic review that describes existing artificial orthography learning experiments and examines how this body of evidence informs our understanding of statistical learning in reading and spelling acquisition. Among other findings, this study showed that most experiments focused on orthography-phonology mappings with native English-speaking participants. Evidence from orthotactic learning experiments suggests that people can become sensitive to statistical regularities after brief periods of exposure. However, this paradigm faces several limitations, particularly concerning ecological validity and the comparability of findings across experiments. Study 2 reports two artificial orthography learning experiments (Experiments 1 & 2) investigating whether native English-speaking adults could simultaneously learn both semantic (where possible spellings depend on grammatical word class) and graphotactic regularities (where possible spellings depend on earlier graphemes) in an artificial lexicon. Participants showed successful generalisation of semantic patterns at post-test, but there was no conclusive evidence of graphotactic learning. Learning was stronger when the semantic patterns were assigned to nouns and verbs compared to adjectives and adverbs. This learning was also associated with participants’ ability to verbalise the patterns at post-test. Study 3 includes two additional artificial orthography learning experiments examining whether participants could learn graphotactic patterns modelled on French noun pluralisation patterns and whether semantic cues facilitated this learning. The artificial orthography was presented in Latin alphabets (Experiment 3) and symbols (BACS-2 fonts; Experiment 4) to further examine whether unintended phonological cues impacted graphotactic learning. In both experiments, participants demonstrated graphotactic learning after brief exposure, although learning was stronger when Latin alphabets were used. Consistent with findings from Study 2, participants who could verbalise the patterns at post-test showed better generalisation. Contrary to predictions, there was limited evidence that semantic cues facilitated graphotactic learning. Combining a systematic review of existing evidence with experimental work, this thesis provides new insights into the learning mechanisms underlying reading and spelling acquisition. This research addresses a knowledge gap in understanding how learners acquire regularities beyond orthography-phonology mappings in written language. It also holds implications for theories of literacy development, spelling instruction and the use of artificial orthography in psycholinguistic research

    Charge recombination in polythiophene: non-fullerene acceptor solar cells with IE offsets exceeding 1 eV

    Get PDF
    In organic solar cells the energetic landscape of the donor–acceptor heterojunction determines the efficiency of charge generation and charge recombination processes, and thereby the device performance. Here, we present a study on a series of 15 donor–acceptor bulk heterojunctions (BHJs) consisting of either the donor polymer poly(3-hexylthiophene) (P3HT) or poly[2,5-bis(3-tetradecylthiophen-2-yl)thieno[3,2-b]thiophene] (pBTTT-C14) and selected non-fullerene acceptors (NFAs), spanning a wide range of interfacial energetics. We demonstrate that the internal quantum efficiency (IQE) is limited by geminate and non-geminate recombination processes and, importantly, decreases with the energy difference between the donor's ionization energy (IE) and the acceptor's electron affinity (EA), in other words, the diagonal bandgap, specifically if less than 1 eV, regardless of the interfacial IE offset. The dependence of charge recombination on the diagonal bandgap can be explained in the framework of the energy gap law. Our results provide further insight into the importance and impact of interfacial energetics in donor:NFA blends with large IE offsets

    Childhood outcomes in children with Hirschsprung disease: a population-based data linkage study in England

    Get PDF
    Objective: Hirschsprung disease (HSCR) is a rare congenital intestinal condition that, despite corrective surgery, can lead to recurrent hospitalisations and reduced quality of life throughout childhood. There is limited population-level evidence on the causes of these admissions or associated educational needs. Design/methods: Using linked health and education data from the Education and Child Health Insights from Linked Data database, we created a cohort of all children born in England between 2002 and 2020. We compared admission rates, mortality, surgical procedures and causes of admission for children with and without HSCR at ages 0–4, 5–9 and 10–14 years. We also assessed the prevalence of recorded Special Educational Needs and Disability (SEND) by Year 1 of primary school (age 6). Results: Among 11 261 227 children, 3227 (0.03%) had HSCR. By age 4, 95.0% of children with HSCR had been readmitted, compared with 40.2% without. Common admission reasons included constipation, gastroenteritis, intestinal infection, abdominal pain and nausea or vomiting. 81.4% of children with HSCR had two or more surgical procedures between ages 0 and 4, compared with 2.9% in children without HSCR. Mortality by age 4 was 3.2% for HSCR versus 0.5% for non-HSCR children. By age 6, 44.0% of children with HSCR had recorded SEND compared with 17.9% of those without. Conclusion: Children with HSCR experience substantially higher hospital readmission rates, more surgeries and greater mortality up to age 14 than their peers. They are also more likely to require educational support, independent of comorbidities such as Down syndrome. Improvements in surgical and long-term care, including novel or complementary approaches, alongside enhanced educational and psychosocial support, are needed to improve outcomes and quality of life across childhood and adolescence

    Model-guided geospatial surveillance system for antimalarial drug resistance

    Get PDF
    Disease surveillance activities are usually resource-constrained and should be optimised to generate the most informative scientific findings, and to make the best use of time, finances, and personnel. India has a high population density, diverse geography and climatic conditions, and difficult terrain. With respect to malaria, Plasmodium falciparum and Plasmodium vivax are endemic, with substantial variability of transmission across the country. While for P. vivax, drug efficacy appears to be homogeneous within the country, for P. falciparum malaria, the drug resistance pattern varies from the northeastern region to the central region. Accounting for these complexities, we develop a decision-making framework guided by geospatial modelling outputs to identify prospective study sites for surveillance of molecular markers of antimalarial drug resistance in P. falciparum malaria in India. We first retrieve existing data on the prevalence of validated markers of resistance to artesunate and sulfadoxine-pyrimethamine from the World Wide Antimalarial Resistance Network (WWARN) Surveyor database. We then incorporate these data into a geostatistical model to estimate the prevalence of these markers across India and identify areas with high median estimated marker prevalence and high uncertainty. Finally, we create an interactive dashboard using the RShiny software package to simplify the process of selecting sites for future molecular surveillance. Our framework helps to ensure that operational decision-making is supported by data and modelling outputs. We demonstrate the utility of our framework by selecting sites for molecular surveillance of P. falciparum malaria in India

    Virtual reality, value, and the external world

    Get PDF
    Nozick famously argued that life inside an Experience Machine (EM) is dismal because it severs our connection to external reality. Would life spent in Virtual Reality (VR) be similarly dismal? Chalmers argues that because VR is importantly different from an EM, life in VR can be roughly as good as life outside. Chalmers further argues for Virtual Realism, the view that virtual worlds can be unqualifiedly real. I examine the relation between these claims about VR and the idea that there’s special value in connecting to the external world. The EM itself, I argue, can only weakly support the value of such a connection. And VR, as understood by Chalmers, is different from the EM precisely by permitting us to retain extensive links to the non-virtual world. We can better test the value of a connection to external reality by considering a more restrictive version of VR. This narrower form of VR is, I argue, inferior to life outside in several ways. Some of my arguments for this also support the idea of a valuable connection with reality—but only if Virtual Realism is false. By contrast, and contrary to the impression given by Chalmers, Virtual Realism seems to make little difference to questions about the value of life in VR vs. life outside

    Parent-led CBT delivered via online and telephone support alongside usual school practice versus usual school practice only for young children identified as at-risk for anxiety disorders through screening in schools: a cluster randomised controlled trial

    Get PDF
    Background: Providing accessible CBT for young children identified as at-risk for anxiety disorders through screening in schools could reduce later problems. This study aimed to evaluate the effectiveness of parent-led CBT delivered via online and telephone call support alongside usual school provision, compared to usual school provision only, for young children identified through screening as having at least one risk. Methods: We conducted a pragmatic, parallel group, superiority cluster randomised controlled trial in 95 primary/infant schools in England. Parents of children (aged 4-7) in sampled classes completed screening, and children who screened positive for one or more risks (anxiety symptoms/inhibition/parent anxiety) were eligible for the trial. Schools (clusters) were randomised (1:1) to intervention or usual school practice, stratified by school-level deprivation. Schools in both arms continued with usual provision, and parents in intervention schools were offered parent-led CBT via online and telephone support. The primary outcome was the presence of an anxiety disorder diagnosis at 12-months, assessed via the ADIS-P administered by independent assessors. Secondary clinical outcomes included parent-reported child anxiety symptoms, related interference, externalising symptoms, additional risks and intervention targets at 12-weeks and 12-months. Primary analyses were conducted on the full intention-to-treat population. The trial was prospectively registered with ISRCTN 82398107. Results: 2328 children were screened; 1172 were eligible; 865 enrolled. 48 schools (434 children) were assigned to intervention and 47 schools (431 children) to usual school practice. At 12-months the overall frequency of anxiety disorders was low, 6.8% (21/310) of children in the intervention arm compared to 11.5% (36/312) in the usual school practice arm; this difference was not statistically significant (adjusted odds ratio 0.67 [0.37 to 1.21], p=0.19). However, the intervention was superior to usual school practice across all secondary outcomes (standardised mean difference: 0.15 to 0.47 at 12-weeks; 0.19 to 0.41 at 12-months). No serious adverse events were reported. Conclusions: Although the intervention did not significantly reduce anxiety disorders at 12-months, improvements across all other assessed outcomes indicate this approach brings wider immediate benefits and reduces known risks for future anxiety disorders. Future research needs to consider longer-term preventative effects

    Parents' experiences in accessing services for their autistic children in the United Kingdom: A meta‐synthesis

    Get PDF
    Background: Parents of autistic children support their children through additional challenges, often experiencing adversity as a result. Such parents report high support needs, yet service provision is often limited. Services often support children through providing various psychological interventions to parents. Quantitative evidence for such interventions is mixed and qualitative evidence is sparse. This review therefore aimed to synthesise the perspectives of UK parents regarding interventions for their autistic child. Method: The databases Scopus, Embase, Medline, PubMed, PsycInfo, CINAHL, Web of Science and ASSIA were searched in February 2025. Inclusion criteria constituted qualitative articles published in English from 2004 onwards exploring UK parents’ perspectives of interventions aimed at supporting autistic children. Articles were evaluated using Standard Quality Assessment Criteria. Thematic meta‐synthesis was conducted. Results: Fourteen papers were identified: eight high‐quality, one medium‐quality, and four low‐quality. Interventions were psychoeducational behavioural, communication‐based, sensory‐related or mental‐health based in nature. Themes included change, relationship with help, parents' need to process and solidarity. Conclusions: Facilitators of positive change included learning, empowerment, structure and rigour, while barriers included delivery issues and unhelpful information. Parents reported finding solidarity amongst similar parents helpful. Reflective space was deemed useful in facilitating new understanding of autistic lives. Methodological quality varied, with more reflexive and theoretically grounded research encouraged. Future research should also consider implementing embedding processes into qualitative designs

    Long-read sequencing reveals increased isoform diversity in key transcription factor effectors of intercellular signalling at the invertebrate-vertebrate transition

    Get PDF
    Several intercellular signalling pathways (namely wingless - Wnt, Hedgehog - Hh, and Bone Morphogenetic Protein - BMP) are used repeatedly in animals throughout development and evolution, and are also frequent targets for disease-associated disruptions. We have previously shown that the major transcriptional effectors of βcatenin-dependent Wnt signalling, the TCF/LEF proteins, in contrast to other pathway components, have a higher gene number and isoform diversity in vertebrates versus invertebrates, but this increased diversity has only been poorly quantified. Considering that isoform diversity correlates with organism complexity, any increase in major signalling effectors is likely to have made a significant contribution to vertebrate evolution. Using de novo long-read transcriptomes, we compared isoform number per gene for the chordates Ciona intestinalis, Lampetra planeri and Xenopus tropicalis, thus encompassing the invertebrate sister group to vertebrates, as well as a cyclostome and a gnathostome vertebrate. Our results implicate an increase in isoform diversity of the transcription factors of major intercellular signalling pathways as having a disproportionate role in the evolutionary origin and diversification of vertebrates

    150,487

    full texts

    324,139

    metadata records
    Updated in last 30 days.
    Oxford University Research Archive is based in United Kingdom
    Access Repository Dashboard
    Do you manage Open Research Online? Become a CORE Member to access insider analytics, issue reports and manage access to outputs from your repository in the CORE Repository Dashboard! 👇