136986 research outputs found
Sort by
Things of Life in Times of Extremes: Survival Materialities During the Soviet Famines in Ukraine
No full textProtein Target Highlights in CASP16: Insights From the Structure Providers
No full textThis article presents an in-depth analysis of selected CASP16 targets, with a focus on their biological and functional significance. The authors highlight the most relevant features of the target proteins and discuss how well these were reproduced in the submitted predictions. While the overall performance of structure prediction methods remains impressive, challenges persist, particularly in modeling rare structural motifs, flexible regions, small molecule interactions, posttranslational modifications, and biologically important interfaces. Addressing these limitations can strengthen the role of structure prediction in complementing experimental efforts and advancing both basic research and biomedical applications
Sexual and social functions of ornamental wing feathers in Black Swans (Cygnus atratus)
No full textStudies on the evolution of ornamentation have traditionally focused on sexual selection, although ornaments may also signal an individual's quality in non‐sexual contexts. Using wild‐captured Black Swans
Cygnus atratus
, we conducted two‐way association trials to study how a putative mutual ornament (curled wing feathers) influenced swans' intersexual and intrasexual preferences. When allowed to freely associate with two candidate birds matched for size and condition, focal males preferentially associated with the candidate with the higher number of curled wing feathers whereas focal females showed no consistent preference. Males preferred the more ornamented bird during both sexual trials (where both candidates were female) and social trials (where both candidates were male), supporting the hypothesis that monomorphic ornaments can function in both sexual and social interactions
Return on investment from interventions to prevent common mental disorders among adolescents in Australia: A modelling study
No full textOBJECTIVE: The prevalence of common mental disorders (CMDs, depression/anxiety) among Australian adolescents is increasing. Evidence is needed to inform how prevention efforts can be focused for greatest impact. This study used a compartmental model to estimate how investments could be optimally allocated across interventions to prevent CMDs in adolescents, and the corresponding health and economic benefits. METHODS: The Australian 0- to 19-year-old population was disaggregated by age (0-9, 10-14, 15-19), sex (male, female) and CMD experience (never, currently, previously). Twelve prevention interventions were included following targeted reviews. The model was calibrated to Australian population survey data and included sociodemographic risk factors from the literature that were relevant for included interventions (bullying, financial stress, poor school connectedness, racial discrimination, child maltreatment). An optimisation estimated how an additional AUD1 billion per annum over 2024-2050 could be allocated across interventions to maximise the number of adolescents turning 20 without a lifetime experience of CMD. RESULTS: The benefit-cost ratios of interventions varied widely depending on whether they impacted CMDs directly or indirectly by reducing a risk factor, and also whether CMD was their primary purpose or not (e.g. poverty interventions). Optimally investing AUD1 billion p.a. over 2024-2050 could lead to 144,577-791,277 fewer adolescents experiencing a CMD by age 20 (AUD18,507/lifetime CMD experience averted), with lower bound benefit-cost ratios ranging from 5.0 to 19.2. Optimally, budget allocations should include combinations of interventions targeting different age groups, sub-populations and risk factors. CONCLUSIONS: Investing in evidence-based prevention interventions for adolescent mental health is likely to have a favourable return-on-investment
Long-Term Follow-Up of Patients in a Prospective Study of NA Discontinuation Identifies Different Patterns of HBsAg Loss
No full textBACKGROUND AND AIMS: Discontinuing nucleos(t)ide analogues (NAs) may lead to functional cure (HBsAg loss) in selected patients with chronic hepatitis B (CHB). We evaluated the rates and predictors of HBsAg loss during long-term follow-up in a prospective cohort. METHODS: This real-world extension study followed participants from a prospective trial of NA discontinuation. All patients had HBeAg-negative CHB without cirrhosis. Efficacy outcomes (including HBsAg loss and decline) and safety outcomes [including hepatitis flare and hepatocellular carcinoma (HCC)] were evaluated. RESULTS: Amongst 97 participants (85% Asian), with a median follow-up of 7 years, the cumulative incidence of HBsAg loss was 10%, 13% and 22% at 5, 7 and 9 years after stopping NA. HBsAg loss was associated with a lower end-of-treatment (EOT) HBsAg level (HR = 0.28, p < 0.001), older age (HR = 1.14, p = 0.005) and peak off-treatment HBV DNA level (OR = 0.50, p = 0.002). Participants with EOT HBsAg level ≤ 10 IU/mL experienced early HBsAg loss (< 96 weeks) without ALT flares whilst those with EOT HBsAg level ≥ 10 IU/mL experienced late (≥ 96 weeks) HBsAg loss, often following ALT flares (5/8 cases). No cases of hepatic decompensation, liver transplantation or death occurred. Median liver stiffness did not increase. HCC was diagnosed in three individuals (4.4/1000 person-years). CONCLUSION: The rate of functional cure increased during long-term follow-up but remained low. EOT HBsAg strongly predicted the likelihood and timing of HBsAg loss. ALT flares were associated with HBsAg decline, and in some cases, with delayed HBsAg loss. TRIAL REGISTRATION: The clinical study was supported by the National Health and Medical Research Council of the study clinical trial ID is NCT02581033
Pictures of power play: Visual language and analysis of Victorian electricity
No full textAustralia’s electricity system is undergoing a transition, with each jurisdiction contributing to this shift. Despite progress, about 70% of Victoria’s electricity is still generated from brown coal, underscoring the importance of understanding its historical development to inform future transitions. This article presents the Socio-Technical Layout (STL) framework as a graphical tool to map, diagnose, and analyse key elements of socio-technical systems, including technologies, actors, markets and energy–money flows. Applying the STL to Victoria’s electricity system uncovers three development phases: emerging, public, and private; and reveals that: (i) complexity — traditionally linked to scale — depends on the lenses (technological, organisational) and viewpoint (household, system operator); and (ii) transformative actors institutionalised socio-technical solutions aligned with societal values and needs. This approach also carries implications for transition agendas. As an explanatory framework with diagnostic capabilities, STLs provide a common language to enhance participatory processes. Stakeholders can use STLs to identify areas ripe for change, test and evaluate potential modifications from different viewpoints, and identify what is needed to support them (new markets, actors). STLs can be formalised into computer models, bridging qualitative and quantitative research. STLs should be complementary to narratives and other transition frameworks, given their representational limitations
Outcome measures in prehabilitation interventions for total hip and knee arthroplasty: A scoping review
No full textObjectiveThis scoping review aimed to map outcome measures collected in randomised controlled trials investigating prehabilitation interventions in total hip and knee arthroplasty, and timepoints of collection.Data sourcesA systematic search of MEDLINE, EMBASE, Web of Science, Cochrane, and CINAHL was conducted.MethodsThis review followed the Joanna Briggs Institute Scoping Review methodology. Outcome domains, concepts of interest and assessment tools were characterised using the International Society for Pharmacoeconomics and Outcomes Research Framework, and timepoints for data collection were extracted.ResultsNinety-two trials (published between June 2001 and March 2025) were included. Most delivered unimodal prehabilitation, with exercise the most common intervention (n = 37). The review identified 36 outcome concepts measured with 219 assessment tools. Patient-reported outcomes were collected in 92% of trials (n = 85), and was the most heterogenous domain with 102 assessment tools. Performance-based outcomes, most commonly muscle strength, were collected in 66% of trials (n = 61) and utilised 47 different tools. Observer-reported outcomes were reported in 60% of trials (n = 55), with healthcare utilisation (e.g. hospital length of stay) the most common concept. Clinician-reported outcomes were reported in 48% of trials (n = 44) and most frequently included post-operative complications. Biomarker outcomes were rare (n = 7, 8%). Timing of outcome collection varied, with just over half the trials collecting both a pre-operative and post-operative timepoint.ConclusionThis review identified significant variability in outcome measures collected in prehabilitation trials for total hip and knee arthroplasty, highlighting the need for a core set of assessments to facilitate consistent reporting and robust meta-analyses of prehabilitation efficacy
Risk aggregation and stochastic dominance for a class of heavy-tailed distributions
No full textAbstract
We introduce a new class of heavy-tailed distributions for which any weighted average of independent and identically distributed random variables is larger than one such random variable in (usual) stochastic order. We show that many commonly used extremely heavy-tailed (i.e., infinite-mean) distributions, such as the Pareto, Fréchet, and Burr distributions, belong to this class. The established stochastic dominance relation can be further generalized to allow negatively dependent or non-identically distributed random variables. In particular, the weighted average of non-identically distributed random variables dominates their distribution mixtures in stochastic order
AI collaboration or cheating? Using explainable authorship verification to measure AI assistance in academic writing
No full textAs human-AI collaboration becomes increasingly prevalent in educational contexts, understanding and measuring the extent and nature of such interactions pose significant challenges. This research investigates authorship verification (AV) techniques to quantify AI assistance in academic writing, focusing on transparency and interpretability. We structured our investigation into three stages: dataset selection and expansion, AV method development, and systematic evaluation. Using three datasets in Stage 1, including PAN-14 and two from University of Melbourne students, we expanded the data to include LLM-generated texts, totalling 1,889 documents and 540 authorship problems from 506 students. Next, we developed an adapted Feature Vector Difference (FVD) authorship verification method to construct academic writing profiles for students, capturing meaningful stylistic features. Lastly, our AV method was evaluated across multiple scenarios including distinguishing between student-authored and LLM-generated texts, and detecting AI mimicry using standard authorship verification metrics such as AUC, c@1, and F1. Results showed that our approach effectively distinguished between student-authored and AI-generated texts, even under mimicry scenarios, offering educators actionable insights into students' writing progress