Cureus. 2025 Mar 27;17(3):e81305. doi: 10.7759/cureus.81305. eCollection 2025 Mar.

ABSTRACT

Attention-deficit hyperactivity disorder (ADHD) is generally treated with stimulant medications without significant complications. Delusional parasitosis (Ekbom syndrome) can occur secondary to ADHD treatment. It is a rare condition defined as having a fixed, false belief that one is infected with insects, parasites, or organisms and that one experiences cutaneous sensations without any clinical evidence of infestation. Although stimulant treatment with methylphenidate or mixed amphetamine salts has been associated with delusional parasitosis, there is yet a case in the literature illustrating delusional infestation secondary to lisdexamfetamine. The following case is unique in that lisdexamfetamine caused delusional parasitosis in a 53-year-old man with ADHD who previously tolerated mixed amphetamine salts and armodafinil without side effects. The discontinuation of lisdexamfetamine, coupled with a second-generation antipsychotic, quickly resolved the delusion. For those who may prescribe lisdexamfetamine or treat patients with ADHD, it is crucial to carefully assess medication use, as discontinuation or dose adjustment of the suspected causative drug can have a positive impact on the course of delusional parasitosis.

PMID:40291316 | PMC:PMC12034334 | DOI:10.7759/cureus.81305

Cureus. 2025 Mar 28;17(3):e81364. doi: 10.7759/cureus.81364. eCollection 2025 Mar.

ABSTRACT

Drug-induced thrombocytopenia (DITP) is a rare but serious immune-mediated reaction characterized by drug-dependent antibodies that bind platelet surface glycoproteins, leading to severe thrombocytopenia. Biologic therapies, including IL-23 inhibitors like risankizumab, have been implicated in such adverse events. We present the case of a 47-year-old male with a history of psoriasis and prior deep vein thrombosis, who developed severe bleeding manifestations shortly after initiating risankizumab therapy for psoriatic arthritis. Clinical evaluation included mucocutaneous bleeding, petechiae, and a precipitous drop in platelet count to 3 x 10^3/uL. Management strategies involved platelet transfusions, high-dose steroids, intravenous immunoglobulin (IVIG), and thrombopoietin receptor agonists due to inadequate initial response. Despite aggressive treatment, the patient's thrombocytopenia persisted, necessitating prolonged hospitalization and consideration of alternative therapies. This case underscores the critical importance of recognizing and managing rare hematologic complications associated with biologic therapies. Vigilance in monitoring platelet counts during IL-23 inhibitor therapy is essential to mitigate severe adverse outcomes.

PMID:40291297 | PMC:PMC12034315 | DOI:10.7759/cureus.81364

Front Public Health. 2025 Apr 11;13:1572611. doi: 10.3389/fpubh.2025.1572611. eCollection 2025.

ABSTRACT

INTRODUCTION: Pharmacovigilance plays a vital role in ensuring drug safety and protecting public health. Oropharyngeal adverse drug reactions (O-ADRs) are found to be under-reported, especially by oral health professionals, limiting the identification and management of these events.

AIMS: This study aimed to evaluate the knowledge, attitudes, and practices (KAP) of healthcare professionals and students regarding O-ADRs and to assess their specific expertise by a self-e-learning test.

MATERIALS AND METHODS: A cross-sectional survey was conducted using a KAP questionnaire between April 2023 and April 2024, involving 943 participants, including physicians, dentists, dental hygienists, and students. Additionally, three sets of self-e-learning tests on O-ADRs were administered. The study employed descriptive statistics, Kruskal-Wallis tests, and logistic regression to analyze factors affecting KAP and reporting behaviors.

RESULTS: Significant gaps in KAP were identified. Only 26.5% of participants demonstrated frequent best practices for reporting O-ADRs, with dentists and dental hygienists showing lower reporting rates (13.8% and 9.3%, respectively) compared to physicians (18.8%). The results of logistic regression analyses showed that practical knowledge was significantly associated with work experience (OR = 2.15, p = 0.026). Students exhibited the lowest levels of practical knowledge and reporting proficiency, with only 17.6% demonstrating competence. The self-e-learning test highlighted knowledge deficits: only 22.9% of participants correctly identified O-ADR associated with antiseptic mouth rinses, additional 30.2% recognized those linked to antimicrobial drugs.

CONCLUSIONS: This study highlights the need for targeted educational interventions to address gaps in O-ADR knowledge and practice. Tailored training, user-friendly digital tools, and a strong pharmacovigilance culture are crucial for improving reporting rates and ensuring patient safety.

PMID:40290502 | PMC:PMC12021887 | DOI:10.3389/fpubh.2025.1572611

In Silico Pharmacol. 2025 Apr 24;13(2):70. doi: 10.1007/s40203-025-00355-z. eCollection 2025.

ABSTRACT

Marburg Virus (MARV) presents a significant threat to human health, highlighting the urgent need for effective therapeutics. The MARV genome encodes a multifunctional 'large' L protein that plays a crucial role in polymerase, capping, and methyltransferase activities. Within this protein, the 2'-O-methyltransferase (2'-O-MTase) domain is essential for viral replication and immune evasion, making it a promising therapeutic target. However, the lack of structural data on this domain limits drug discovery efforts. To address this challenge, we utilized AlphaFold2 to predict a 3D structure of the MARV 2'-O-MTase domain. Molecular docking with its natural ligand, S-adenosyl methionine (SAM), allowed us to identify key active-site residues involved in ligand binding. We then screened 62 known inhibitors against this domain and identified four promising candidates: Lifirafenib (- 9.5 kcal/mol), Dolutegravir (- 8.5 kcal/mol), BRD3969 (- 8.3 kcal/mol), and JFD00244 (- 8.2 kcal/mol). Further, we assessed the pharmacokinetic and pharmacodynamic properties of these compounds to evaluate their drug-likeness. Molecular dynamics simulations, along with MM/GBSA free energy calculations, confirmed stable interactions between the selected inhibitors and the target domain. While these findings highlight promising candidates for MARV, experimental validation through in vitro and in vivo assays is essential to assess their safety and efficacy.

SUPPLEMENTARY INFORMATION: The online version contains supplementary material available at 10.1007/s40203-025-00355-z.

PMID:40291443 | PMC:PMC12018677 | DOI:10.1007/s40203-025-00355-z

ACS Omega. 2025 Feb 22;10(15):14980-14993. doi: 10.1021/acsomega.4c10338. eCollection 2025 Apr 22.

ABSTRACT

Huntington's disease (HD) and Spinocerebellar Ataxia (SCA) are debilitating neurological disorders triggered by the expansion of CAG sequences within the specific genes (HTT and ATXN, respectively). These are characterized as poly glutamine (polyQ) disorders, which are marked by widespread neurodegeneration and metabolic irregularities across systemic, cellular, and intracellular levels. This study aimed to identify small molecules that specifically interact with and target the toxic CAG repeat RNA. Here, we investigated the neuroprotective effects of Oseltamivir, an antiviral drug, against the HD and SCA-causing CAG repeats, through biophysical, cellular, and Drosophila model-based studies. Using a multidimensional approach encompassing biophysical techniques, cellular assays, and a Drosophila model, we explored Oseltamivir's interaction with toxic CAG repeat RNA. Our comprehensive analyses, including circular dichroism (CD), isothermal titration calorimetry (ITC), electrophoretic mobility shift assay (EMSA), and nuclear magnetic resonance (NMR) spectroscopy, demonstrated Oseltamivir's specific binding affinity for AA mismatches and its potential to mitigate the toxicity associated with polyQ aggregation. Moreover, the identified U.S. FDA-approved drug effectively mitigated polyQ-induced toxicity in both HD cells and the Drosophila model of the disease. The results obtained from this drug repurposing approach are indicative of the neuro-shielding role of Oseltamivir in HD and several SCAs, paving the way for its translation into clinical practice to benefit patients afflicted with these devastating diseases.

PMID:40290909 | PMC:PMC12019426 | DOI:10.1021/acsomega.4c10338

EClinicalMedicine. 2025 Mar 25;82:103102. doi: 10.1016/j.eclinm.2025.103102. eCollection 2025 Apr.

ABSTRACT

The kidney disease: Improving Global Outcomes (KDIGO) guideline recommends assessing kidney function using glomerular filtration rate (GFR) either through direct measurement or through estimation (eGFR) and describes a standardised classification of reduced kidney function. KDIGO guidelines have been adopted by most internal medicine specialities for the assessment and classification of kidney function, but not by cancer medicine. The development of the International Consensus Guideline on Anticancer Drug Dosing in Kidney Dysfunction (ADDIKD) aims to overcome the perceived challenges with KDIGO recommendations by describing their utility in patients with cancer. Two virtual, consensus building workshops were held consecutively, involving international, multidisciplinary participants (Part 1 of ADDIKD development). During these workshops, three consensus recommendations were agreed upon based on KDIGO's principles; to standardise kidney function assessment, classify kidney function, and determine a uniform approach to dose anticancer drugs in patients with reduced kidney function. Cancer clinicians attending the workshops identified issues regarding the adoption of KDIGO's recommendations. These issues were addressed by nephrologists, clinical pharmacologists, and other clinicians with extensive experience in the contemporary assessment of kidney function. The key concern for cancer specialists was a hesitancy to move away from the familiar and long-standing practice of using the Cockcroft-Gault equation to estimate creatinine clearance. The consensus building within the two multidisciplinary workshops allowed a thorough assessment of the evidence and clarified how directly measured GFR and eGFR, rather than creatinine clearance, could be optimally utilised in cancer care. The development of Part 1 of the ADDIKD guideline represents a standardised, contemporary approach to the assessment, classification, and utility of kidney function in the setting of cancer care and it harmonises with the approach used in other areas of medicine internationally.

FUNDING: Development of the ADDIKD guideline is funded by the Cancer Institute NSW as part of the NSW Government and received no funding from external commercial sources.

PMID:40290845 | PMC:PMC12034077 | DOI:10.1016/j.eclinm.2025.103102

EClinicalMedicine. 2025 Mar 25;82:103161. doi: 10.1016/j.eclinm.2025.103161. eCollection 2025 Apr.

ABSTRACT

Part 2 of the International Consensus Guideline on Anticancer Drug Dosing in Kidney Dysfunction (ADDIKD) offers drug-specific consensus recommendations based on both evidence and practical experience. These recommendations build upon the kidney function assessment and classification guidelines established in Part 1 of ADDIKD. Here we illustrate how dosing recommendations differ between ADDIKD and existing guidance for four commonly used drugs: methotrexate, cisplatin, carboplatin and nivolumab. We then describe how the recommendations can be distilled into practice points for methotrexate and cisplatin. While ADDIKD is a significant improvement from previous guidelines, adoption of this new guideline requires further endorsement from key external stakeholders, 'change championing' by clinicians locally and encouraging its integration into existing reference sources, clinical trial protocols and electronic prescribing systems.

FUNDING: Development of the ADDIKD guideline is funded by the NSW Government as part of the Cancer Institute NSW and received no funding from external commercial sources.

PMID:40290844 | PMC:PMC12034076 | DOI:10.1016/j.eclinm.2025.103161

Bioinform Biol Insights. 2025 Apr 12;19:11779322251328269. doi: 10.1177/11779322251328269. eCollection 2025.

ABSTRACT

Drug repositioning holds great promise for reducing the time and cost associated with traditional drug discovery, but it faces significant challenges related to data imbalance and noise in negative samples. In this article, we introduce a novel method leveraging high negative oversampling (HNO) to address these challenges. Our approach integrates HNO with advanced techniques such as network-based graph mining, matrix factorization, and Bayesian inference, specifically designed for imbalanced data scenarios. Constructing high-quality negative samples is crucial to mitigate the detrimental effects of noisy negative data and enhance model performance. Experimental results demonstrate the efficacy of our approach in enhancing the performance of drug discovery models by effectively managing data imbalance and refining the selection of negative samples. This methodology provides a robust framework for improving drug repositioning, with potential applications in broader biomedical domains.

PMID:40290635 | PMC:PMC12033409 | DOI:10.1177/11779322251328269

Acta Derm Venereol. 2025 Apr 28;105:adv40697. doi: 10.2340/actadv.v105.40697.

ABSTRACT

In recent years, molecular target drugs have become integral in treating malignant tumours. Multikinase inhibitors (MKIs) have been associated with serious skin disorders, including hand-foot skin reaction (HFSR), which impair patient quality of life, often disrupting activities of daily living necessitating dose reduction or discontinuation. As the pathogenic mechanisms of these skin disorders are unknown, no effective treatments have been established. Previously, by drug repurposing using an in vitro culture system, certain azole antifungal drugs (AFDs) were identified that prevented sorafenib-induced cell death of normal human epidermal keratinocytes. In this study, topical ketoconazole demonstrated clinical improvement in hyperkeratosis and pain associated with sorafenib-induced HFSR. Investigation of the mechanism using the in vitro culture system revealed sorafenib to be particularly cytotoxic among MKIs. Annexin V and TUNEL staining revealed apoptosis was mainly involved in this cytotoxicity. Antibody arrays and western blot showed increased levels of secretion of interleukin-1 receptor antagonist and macrophage migration inhibitory factor in culture supernatants. AFDs suppressed the secretion of these cytokines and reduced apoptosis in keratinocytes. This study reveals one aspect of the pathogenesis of sorafenib-induced HFSR and demonstrates that AFDs may be an effective treatment.

PMID:40289816 | DOI:10.2340/actadv.v105.40697

J Mater Chem B. 2025 Apr 28. doi: 10.1039/d5tb00365b. Online ahead of print.

ABSTRACT

The rapid advancement of implantable biomedical materials necessitates a comprehensive understanding of macrophage interactions to optimize implant immunocompatibility. Macrophages, key immune regulators, exhibit phenotypic plasticity by polarizing into pro-inflammatory (M1) or anti-inflammatory (M2) subtypes. Conventional phenotyping techniques, such as flow cytometry and immunostaining, provide insights but have limitations related to fixation and endpoint analysis. This study presents a high-throughput, label-free macrophage phenotyping approach integrating AI-driven image classification with quantitative phase imaging (QPI). THP-1-derived macrophages were differentiated into M0, M1, M2a, and M2c phenotypes, and their morphological and refractive index properties were analyzed using QPI. Although QPI alone could not fully distinguish phenotypes, deep learning models, including GoogLeNet, ShuffleNet, VGG-16, and ResNet-18, were evaluated, with ResNet-18 achieving over 90% accuracy. Additionally, macrophage responses to collagen coatings (types I, III, and IV) were assessed using machine learning-based phenotyping and cytokine profiling. Collagen I induced an M1 response, collagen III supported a balanced M1/M2 profile, and collagen IV promoted a controlled immune environment. These findings demonstrate the potential of AI-driven QPI as a non-invasive tool for macrophage characterization, offering insights into biomaterial immunocompatibility and informing implant design strategies.

PMID:40289902 | DOI:10.1039/d5tb00365b

NMR Biomed. 2025 Jun;38(6):e70038. doi: 10.1002/nbm.70038.

ABSTRACT

Accurate quantification of metabolites in magnetic resonance spectroscopy (MRS) is challenged by low signal-to-noise ratio (SNR), overlapping metabolites, and various artifacts. Particularly, unknown and unparameterized baseline effects obscure the quantification of low-concentration metabolites, limiting MRS reliability. This paper introduces wavelet analysis-based neural decomposition (WAND), a novel data-driven method designed to decompose MRS signals into their constituent components: metabolite-specific signals, baseline, and artifacts. WAND takes advantage of the enhanced separability of these components within the wavelet domain. The method employs a neural network, specifically a U-Net architecture, trained to predict masks for wavelet coefficients obtained through the continuous wavelet transform. These masks effectively isolate desired signal components in the wavelet domain, which are then inverse-transformed to obtain separated signals. Notably, an artifact mask is created by inverting the sum of all known signal masks, enabling WAND to capture and remove even unpredictable artifacts. The effectiveness of WAND in achieving accurate decomposition is demonstrated through numerical evaluations using simulated spectra. Furthermore, WAND's artifact removal capabilities significantly enhance the quantification accuracy of linear combination model fitting. The method's robustness is further validated using data from the 2016 MRS Fitting Challenge and in vivo experiments.

PMID:40289522 | DOI:10.1002/nbm.70038

Expert Rev Respir Med. 2025 Apr 27. doi: 10.1080/17476348.2025.2499651. Online ahead of print.

ABSTRACT

INTRODUCTION: Idiopathic pleuroparenchymal fibroelastosis (iPPFE) is a rare progressive interstitial lung disease characterized by upper-lobe fibrosis, severe restrictive impairment, and poor prognosis. Unlike idiopathic pulmonary fibrosis, in which acute exacerbations, chronic respiratory failure, and lung cancer are the major causes of death, iPPFE primarily leads to progressive respiratory failure, often complicated by malnutrition and recurrent pneumothorax. Despite growing recognition, its pathogenesis remains unclear and no effective treatments exist.

AREAS COVERED: This review summarizes the epidemiological, clinical, radiological, and pathological features of iPPFE, as well as diagnostic and prognostic advancements. Key prognostic factors include weight loss, reduced forced vital capacity, hypercapnia, and lower-lobe interstitial pneumonia. Serum biomarkers (e.g. latent transforming growth factor-beta binding protein-4) are being explored for early detection and prognostic purposes. Although antifibrotic agents show limited efficacy, supportive care - pulmonary rehabilitation, nutritional management, and pneumothorax prevention - remains essential. Research on the fibroelastotic pathways may inform the development of future therapies.

EXPERT OPINION: IPPFE remains a challenging disease. Therefore, early diagnosis and comprehensive management of this condition are crucial. Future research should refine prognostic models and explore novel therapeutic approaches for treating fibroelastosis. Lung transplantation may be an option for select patients. Further studies are required to optimize these outcomes.

PMID:40289399 | DOI:10.1080/17476348.2025.2499651

Br J Clin Pharmacol. 2025 Apr 27. doi: 10.1002/bcp.70057. Online ahead of print.

ABSTRACT

AIMS: Pharmacogenomics enables personalization of drug therapy improving effectiveness and/or safety. Dihydropyrimidine dehydrogenase [DPYD] testing prior to fluoropyrimidine chemotherapy was commissioned by NHS England in response to an update from the Medicines and Healthcare products Regulatory Agency.

METHODS: A questionnaire developed and approved by the National DPYD Working Group investigated processes of testing, receiving and responding to results. The survey was distributed to Genomics Medicine Service Alliance (GMSA) Clinical Directors and Pharmacy Senior Responsible Officers for dissemination in their geography. Data were collected June-September 2021. All hospitals delivering fluoropyrimidines across the UK were invited to participate.

RESULTS: 131/138 (94.9%) organizations reported testing all patients receiving fluoropyrimidines. In England 76.7% of hospitals sent samples to centrally commissioned genomics laboratories. In the devolved nations, 73.9% sent samples to regional genomics laboratories. Multidisciplinary staff including oncologists, independent non-medical prescribers, clinical nurse specialists, screening pharmacists and chemotherapy nurses requested the test, checked and actioned the result. Self-reported turnaround times varied from <2 days where a regional laboratory was colocated with a chemotherapy centre to >10 days.

CONCLUSION: Through multiprofessional, national collaboration, this is the first report studying the large-scale rollout of a nationally commissioned pharmacogenetic test. Whilst DPYD testing has been successfully implemented, there is a need to standardize and improve end-to-end turnaround times. This has led to the development of a best practice pathway. Critically, GMSAs must build on this implementation to deliver its priorities, in supporting equitable access to future pharmacogenomic testing across a wider cohort of therapies.

PMID:40289265 | DOI:10.1002/bcp.70057

Methods. 2025 Apr 25:S1046-2023(25)00109-4. doi: 10.1016/j.ymeth.2025.04.012. Online ahead of print.

ABSTRACT

Drug-target interaction (DTI) prediction is critical in drug repositioning and discovery. In current metapath-based prediction methods, attention mechanisms are often used to differentiate the importance of various neighbors, enhancing the model's expressiveness. However, in biological networks with small-scale imbalanced data, attention mechanisms are prone to interference from noise and missing data, leading to instability in weight learning, reduced efficiency, and an increased risk of overfitting. To address these issues, we propose the use of average aggregation to mitigate noise, simplify model complexity, and improve stability. Specifically, we introduce a simplified mean aggregation method for DTI prediction. This approach uses average aggregation, effectively reducing noise interference, lowering model complexity, and preventing overfitting, making it especially suitable for current biological networks. Extensive testing on three heterogeneous biological datasets shows that SNADTI outperforms 12 leading methods across two evaluation metrics, significantly reducing training time and validating its effectiveness in DTI prediction. Complexity analysis reveals that our method offers a substantial computational speed advantage over other methods on the same dataset, highlighting its enhanced efficiency. Experimental results demonstrate that SNADTI excels in prediction accuracy, stability, and reproducibility, confirming its practicality and effectiveness in DTI prediction.

PMID:40288620 | DOI:10.1016/j.ymeth.2025.04.012

Farm Hosp. 2025 Apr 26:S1130-6343(25)00009-1. doi: 10.1016/j.farma.2025.02.002. Online ahead of print.

ABSTRACT

OBJECTIVE: Inhaled therapy is essential in cystic fibrosis; however, inhalers have a significant environmental impact due to the greenhouse gases (GHGs) emitted. The environmental impact of a product is estimated by its carbon footprint (CF). Pressurized metered-dose inhalers (pMDIs) have a higher CF than dry powder inhalers (DPIs) and soft mist inhalers (SMIs) due to the incorporation of GHGs. The objectives are to analyze the consumption of inhalers (β2-adrenergic agonist bronchodilators, anticholinergics, and/or corticosteroids) in a cystic fibrosis unit and estimate the generated CF.

METHOD: Retrospective determination (January 2018-December 2023) of consumption and CF (tCO2eq) by type of inhaler was conducted. Consumption and CF trends were evaluated using linear regression.

RESULTS: Annually, 1.529 (1.279-1.613) pMDIs, 1.055 (855-1.333) DPIs, and 28 (20-42) SMIs were dispensed, representing 55.97%, 42.33%, and 1.70%, respectively. A statistically significant positive trend in the consumption of SMIs was observed. The median annual CF was: pMDIs 38.3 (31.2-40.3) tCO2eq, DPIs 0.8 (0.6-0.9) tCO2eq, and SMIs 0.02 (0.02-0.03) tCO2eq, representing 97.86%, 2.04%, and 0.10%, respectively.

CONCLUSIONS: pMDIs were the inhalers with the highest consumption and CF, although their consumption appears to be decreasing, with an increase in the consumption of SMIs.

PMID:40288920 | DOI:10.1016/j.farma.2025.02.002

Microb Pathog. 2025 Apr 25:107628. doi: 10.1016/j.micpath.2025.107628. Online ahead of print.

ABSTRACT

BACKGROUND: Cystic fibrosis (CF) lung disease begins early, and prophylactic antibiotics have been used to prevent Staphylococcus aureus infection. This study examined the lung microbiome in two countries with differing antibiotic practices and its relationship to lung function in young children with CF.

METHODS: A binational, longitudinal, observational study was performed to define the lower airway microbiome in infants with CF. 16S rRNA sequencing was performed using lavage fluid to characterize the lung microbiota in 45 infants with and without prophylactic antibiotic therapy at an average age of approximately 3 months and 14 months. The association between pulmonary function, bacterial community diversities, and taxa was assessed.

RESULTS: Expected CF bacterial genera and non-traditional bacteria, such as Streptococcus, were identified as core taxa. Microbial community shifts were observed in infants who received antibiotic prophylaxis, with lower alpha diversity (ANOVA, P<0.05) and a higher proportion of Streptococcus at the first visit. Beta diversity (FEV0.5z; MiRKAT, P<0.05) and Streptococcus were associated with FEV0.5z (LASSO and linear regression, β<0). Functional annotation suggested that alteration of lung microbiota may be linked to antimicrobial resistance.

CONCLUSIONS: Lung microbial diversity in infants with CF varied between the two countries, particularly during early infancy. A shift in the lung microbiome toward a higher relative abundance of Streptococcus was associated with reduced pulmonary function.

PMID:40288428 | DOI:10.1016/j.micpath.2025.107628

Sleep Med. 2025 Apr 23;131:106534. doi: 10.1016/j.sleep.2025.106534. Online ahead of print.

ABSTRACT

Noninvasive ventilation (NIV) is widely used in children. Only a few devices are life support ventilators. The pressure support (PSV) mode is the most common used mode for home NIV, while assist-control pressure ventilation (PAC) is usually used in patients with abnormal central drive. Patient-ventilator asynchrony (PVA) is common during NIV and may have different causes, such as unintentional leaks, inadequate settings or misunderstanding of the settings. However, PVA may also be due to issues related to the NIV device, which is less common and is challenging. We report here the cases of 5 children with PVA due to trigger issues with a recent life support device.

PMID:40288254 | DOI:10.1016/j.sleep.2025.106534

Sci Rep. 2025 Apr 27;15(1):14732. doi: 10.1038/s41598-025-00115-1.

ABSTRACT

Accurate medium- to long-term runoff forecasting is crucial for flood control, drought resilience, water resources development, and ecological improvement. Traditional statistical methods struggle to utilize multifaceted variable information, leading to lower prediction accuracy. This study introduces two innovative coupled models-SRA-SVR and SRA-MLPR-to enhance runoff prediction by leveraging the strengths of statistical and deep learning approaches. Stepwise Regression Analysis (SRA) was employed to effectively handle high-dimensional data and multicollinearity, ensuring that only the most influential predictive variables were retained. Support Vector Regression (SVR) and Multi-Layer Perceptron Regression (MLPR) were chosen due to their strong adaptability in capturing nonlinear relationships and extracting latent hydrological patterns. The integration of these methods significantly improves prediction accuracy and model stability. By integrating 80 atmospheric circulation indices as teleconnection variables, the models tackle critical challenges such as high-dimensional data, multicollinearity, and nonlinear hydrological dynamics. The Yalong River Basin, characterized by complex hydrological processes and diverse climatic influences, serves as the case study for model validation. The results show that: (1) Compared to baseline single models, the SRA-MLPR model reduced RMSE (from 798.47 to 594.45) by 26% and MAPE (from 34.79 to 22.90%) by 34%, while achieving an NSE (from 0.67 to 0.76) improvement of 13%, particularly excelling in extreme runoff scenarios. (2) The inclusion of teleconnection indices not only enriched the predictive feature set but also improved model stability, with the SRA-MLPR demonstrating enhanced capability in capturing latent nonlinear relationships. (3) A one-month lag in atmospheric circulation indices was identified as the optimal predictor for basin-scale runoff, providing actionable insights into temporal runoff dynamics. (4) To enhance model interpretability, SHAP (SHapley Additive exPlanations) analysis was employed to quantify the contribution of atmospheric circulation indices to runoff predictions, revealing the dominant climate drivers and their nonlinear interactions. The results indicate that the Northern Hemisphere Polar Vortex and the East Asian Trough exert significant control over runoff dynamics, with their influence modulated by large-scale climate oscillations such as the North Atlantic Oscillation (NAO) and Pacific Decadal Oscillation (PDO). (5) The models' scalability is validated through their modular design, allowing seamless adaptation to diverse hydrological contexts. Applications include improved flood forecasting, optimized reservoir operations, and adaptive water resource planning. Furthermore, the study demonstrates the potential of coupled models as generalizable tools for hydrological forecasting in basins with varying climatic and geographic conditions. This study highlights the potential of coupled models as robust and generalizable tools for hydrological forecasting across diverse climatic and geographic conditions. By integrating atmospheric circulation indices, the proposed models enhance runoff prediction accuracy and stability while offering valuable insights for flood prevention, drought mitigation, and adaptive water resource management. These methodological advancements bridge the gap between statistical and deep learning approaches, providing a scalable framework for accurate and interpretable hydrological, climatological, and environmental predictions. Given the escalating challenges brought about by climate change, the findings of this study make contributions to sustainable water management, interpretable decision-making support, and disaster preparedness at a global level.

PMID:40289219 | DOI:10.1038/s41598-025-00115-1

Sci Rep. 2025 Apr 27;15(1):14718. doi: 10.1038/s41598-025-99779-y.

ABSTRACT

Intelligent knowledge about Capsicum annuum L. germplasm could lead to effective management of germplasm. Here, 29 accessions of sweet pepper were investigated in two separate randomized complete block design with three replications in the field condition. Fruit yield accompanied by 13 agro-morphological traits were recorded in two experiments. Genomic fingerprinting of accessions was done by using 10 ISSR primers. The convolutional neural network (CNN) models via outputs of both correlation coefficients and stepwise regression showed the high accuracy of CNN model through correlation coefficients (R2 = 0.879) in predicting fruit yield of sweet pepper. Fruit thickness and fruit width were identified simultaneously as significant components in both models. Genomic best linear unbiased prediction through 65 amplified ISSR loci showed positive and high value of additive gene effect as breeding value for traits identified in the deep learning models. Among studied germplasm, G12, G13, G14, and G25 with positive and high value of breeding value especially for traits constructed the CNN models, recognized as superior genotypes. Regardless of breeding value, multi-trait genotype-ideotype distance index by utilizing all recorded agro-morphological traits simultaneously, revealed G11, G12, G13, and G15 as promising genotypes. So, G12 and G13 which have ideal values of studied traits simultaneously and also positive breeding value could be considered as promising parents for future breeding programs. The study concludes that a CNN model focusing on morphological traits with additive genetic control, combined with the MTSI index, can effectively enhance parental selection in sweet pepper.

PMID:40289216 | DOI:10.1038/s41598-025-99779-y

Sci Rep. 2025 Apr 27;15(1):14733. doi: 10.1038/s41598-025-99777-0.

ABSTRACT

This study aims to effectively improve the quality evaluation system of engineering practice teaching in colleges and universities and enhance the efficiency of teaching management. A brand-new teaching evaluation model is constructed based on the Internet of Things (IoT) technology, combined with complex network analytic hierarchy process and deep learning method. Firstly, with the help of open online course data, Natural Language Processing (NLP) technology and Generative Adversarial Network (GAN) algorithm are used to extract discipline-related features from the course content, and the data of 500 students in 10 majors are simulated and generated. Then, the real university curriculum content, teaching resources, and virtual student data are organically integrated, and two deep learning algorithms, Recurrent Neural Network (RNN) and Convolutional Neural Network (CNN), are introduced. RNN is used to capture time series information, and CNN is used to extract spatial features. Through the hierarchical analysis of complex network, the relationship between different teaching elements is revealed and the hierarchical structure is constructed. Meanwhile, dynamic characteristics are introduced, and continuous model updating and adaptation are realized by randomly combining data to adapt to the changes of actual educational environment. After the course training, data indicators such as students' homework, projects and exams are comprehensively extracted, and the correlation analysis between students' performance and characteristics, time series analysis, feature fusion and weight analysis, model performance evaluation and prediction analysis are carried out. Through the correlation analysis between students' performance and characteristics, the important characteristics that affect learning results are excavated. Time series analysis reveals the changing trend of learning process and better grasps students' learning state. Feature fusion and weight analysis comprehensively consider multiple key features to quantify students' comprehensive performance under different parameter characteristics. Model performance evaluation and prediction analysis compare the prediction results of the model with the actual performance to evaluate the accuracy and stability of the model. The results show that there is a positive correlation between curriculum dependence and interdisciplinary impact index (r = 0.725). The performance of student 3 is relatively stable, with the highest score of 91, and the score of students 7 fluctuates the most, from the lowest 47.9 to the highest 50.2. CNN characteristic index and RNN characteristic index are between 0.18 and 0.78. The comprehensive accuracy of the model in predicting students' actual grades reaches 76-98%, and the prediction consistency varies from 76 to 98%. This study aims to help reveal the relationship between students' performance and teaching evaluation factors, deepen the understanding of the evaluation model of engineering practice teaching in colleges and universities, and provide valuable guidance for optimizing teaching.

PMID:40289170 | DOI:10.1038/s41598-025-99777-0