Transcend Engineering (United States)

Pollution derived from hazardous chemicals and waste has emerged as a critical yet comparatively underexamined driver of global biodiversity loss. Increasing evidence demonstrates that exposure to persistent organic pollutants, heavy metals, pesticides, mismanaged waste, etc., adversely affects species survival, ecosystem function, and genetic diversity. However, the scientific and policy responses to these stressors remain fragmented across multilateral environmental agreements (MEAs) and global governance processes. This assessment studies the role of the Basel, Rotterdam and Stockholm (BRS) Conventions, the Global Framework on Chemicals (GFC), and other international agreements in supporting the implementation of the Kunming–Montreal Global Biodiversity Framework (GBF), with a focus on mechanisms to prevent, monitor and mitigate pollution impacts on biodiversity.The analysis draws upon the 2025 joint BRS-GFC report on their contributions to the GBF targets, a comparative policy mapping against GBF indicators and ongoing collaborative work between biodiversity and chemicals and waste MEAS and International frameworks. In parallel, current work under the GFC’s measurability process is reviewed, examining proposed methodological approaches and indicator frameworks designed to quantify biodiversity benefits arising from improved chemicals and waste management. Particular attention is given to integrative approaches for identifying priority chemicals of concern, assessing exposure pathways and ecosystem-level impacts, and aligning regulatory and technical interventions across multiple actors.Findings indicate that these international frameworks directly support the GBF implementation, particularly interventions related to mitigating pollution impacts, promoting sustainable production and consumption, enhancing knowledge and capacity for informed decision-making, and management and conservation efforts. Moreover, the emerging measurability framework under the GFC provides a basis for developing harmonized global indicators capable of capturing reductions in chemical pressures and associated positive biodiversity outcomes. Furthermore, collaboration with the Convention on Biological Diversity (CBD) demonstrates the potential for cross-regime policy coherence through shared guidance development, integrated technical assistance, and coordinated negotiation inputs.This assessment underscores the importance of integrating chemicals and waste governance into biodiversity policy processes and highlights opportunities for improved scientific collaboration, data harmonization, and implementation coordination. Strengthened alignment across MEAs, reinforced by coherent indicators and shared policy instruments, may significantly enhance the effectiveness of global efforts to halt and reverse biodiversity loss driven by pollution.

Discovering new and viable therapies for genetic diseases is a time-consuming and cost-intensive process, especially for rare disorders. In this study, we highlight how a high-throughput drug discovery platform was utilized to uncover drugs at scale that normalized the signature for a rare neurological neurodevelopmental disease, 19q12 autism spectrum disorder (ASD) associated with deficiencies in ZNF536 and TSHZ3. We first identified the transcriptomic fingerprint of the disease in an in vitro disease model in the form of dysregulated pathways. Subsequently, we measured the biological impact of small molecule drugs in a relevant wild-type cell line and uncovered an approved drug Entrectinib that induced the opposite effect to that in the disease fingerprint, demonstrating the capability to normalize the disease fingerprint. Entrectinib was further prescribed off-label to the identified patient with 19q12 and drug effect was characterized both from blood collection and neuropsychological assessments. Biomarkers from blood recapitulated Entrectinib’s pharmacodynamic effect and normalized the disease signature. We show how generation of transferrable transcriptomics-derived disease signatures allows for measuring drug effects on a signature in related wild-type cell lines, making the screen universally applicable and reducing the need for expensive screens in disease models.Supplementary InformationThe online version contains supplementary material available at 10.1038/s41598-025-26015-y.

This design-based research study examined the implementation fidelity and effectiveness of the SIMPLE Approach, a constructivist intervention integrating the bar model method with metacognitive scaffolding to enhance word problem-solving skills in Kosovo’s lower secondary classrooms. The study aims to address the research gap in that intervention studies in design-based research recognise the critical role of teachers but provide little insight into their actual work and the fidelity of their implementation. To address this research gap, the study investigated (1) alignment of the teachers’ implementation of the SIMPLE Approach and (2) teachers' and pupils' perceptions. Data from classroom observations, teacher diaries, and interviews revealed varying fidelity levels, improving over time. Teacher 1 showed the lowest fidelity, struggling with the bar model method and limiting group work, while Teacher 3 demonstrated the highest fidelity, effectively incorporating all aspects of the SIMPLE Approach and promoting active pupil engagement. Pupils reported initial difficulty with the bar model but ultimately expressed positive perceptions and a desire for continued use. While generalisability is limited, the findings suggest the SIMPLE Approach holds promise but highlights the need for comprehensive teacher training and ongoing support, particularly when introducing novel methods to ensure effective implementation and high fidelity.

In 2023, OpenAI's Democratic Inputs program funded 10 teams to design procedures for public participation in generative AI. In this perspective, we review the results of the project, drawing on interviews with some of the teams and our own experiences conducting participation exercises. We identify several shared yet largely unspoken assumptions of the project and encourage alternative forms of participation in AI perhaps coming from outside the tech industry.

We introduce Dynamic Dropout, a novel regularization technique designed to enhance the training efficiency of Transformer models by dynamically adjusting the dropout rate based on training epochs or validation loss improvements. This approach addresses the challenge of balancing regularization and model capacity, which is crucial for achieving fast convergence and high performance. Our method involves modifying the GPT model to accept a variable dropout rate and updating dropout layers during training using schedules such as linear decay, exponential decay, and validation loss-based adjustments. Extensive experiments on the Shakespeare char dataset demonstrate that Dynamic Dropout significantly accelerates training and improves inference efficiency compared to a baseline model with a fixed dropout rate. The validation loss-based adjustment schedule provided the best overall performance, highlighting the potential of Dynamic Dropout as a valuable technique for training large-scale Transformer models.

Although literature regarding HIV prevention among transgender and nonbinary (TNB) people has grown considerably, important gaps remain, particularly in relation to the HIV pre-exposure prophylaxis (PrEP) care cascade. Additional research is needed to understand when and why TNB people exit the PrEP care cascade to inform interventions to better support these populations. Moreover, most studies have focused on transgender women, though transgender men and nonbinary people also experience a disproportionate prevalence of HIV relative to cisgender populations. This study addresses these gaps by exploring engagement in the PrEP care cascade among a statewide convenience sample of 659 transgender women, transgender men, and nonbinary people, and analyzing how gender identity impacts participant likelihood to continue along the cascade. Data come from the 2018 Michigan Trans Health Study and include self-report data on demographics, sexual behavior, and PrEP awareness, information receipt, referral, initiation, and retention. Participants who self-reported vaginal/front hole or anal sex and did not use barriers (n = 318) were considered eligible for PrEP. Only 21.13% of those eligible for PrEP received information about PrEP, 8.18% were referred to PrEP, and 1.57% initiated PrEP use. A significantly greater proportion of transfeminine and participants who identified with multiple/other genders reported receiving PrEP information than those who identified as transmasculine or nonbinary [χ2 (3, n = 311) =11.34, p = .01]. No other significant gender differences were observed. Providers serving TNB individuals must recognize the diversity of individuals who qualify for information on PrEP, provide trans-affirming care when offering PrEP prescriptions, and consider situating PrEP opportunities in TNB serving spaces to reduce access barriers.