PLoS One. 2025 Mar 10;20(3):e0317630. doi: 10.1371/journal.pone.0317630. eCollection 2025.

ABSTRACT

Portable and wearable electrocardiogram (ECG) devices are increasingly utilized in healthcare for monitoring heart rhythms and detecting cardiac arrhythmias or other heart conditions. The integration of ECG signal visualization with AI-based abnormality detection empowers users to independently and confidently assess their physiological signals. In this study, we investigated a novel method for visualizing ECG signals using polar transformations of short-time Fourier transform (STFT) spectrograms and evaluated the performance of deep convolutional neural networks (CNNs) in predicting atrial fibrillation from these polar transformed spectrograms. The ECG data, which are available from the PhysioNet/CinC Challenge 2017, were categorized into four classes: normal sinus rhythm, atrial fibrillation, other rhythms, and noise. Preprocessing steps included ECG signal filtering, STFT-based spectrogram generation, and reverse polar transformation to generate final polar spectrogram images. These images were used as inputs for deep CNN models, where three pre-trained deep CNNs were used for comparisons. The results demonstrated that deep learning-based predictions using polar transformed spectrograms were comparable to existing methods. Furthermore, the polar transformed images offer a compact and intuitive representation of rhythm characteristics in ECG recordings, highlighting their potential for wearable applications.

PMID:40063554 | DOI:10.1371/journal.pone.0317630

IEEE Trans Vis Comput Graph. 2025 Mar 10;PP. doi: 10.1109/TVCG.2025.3549850. Online ahead of print.

ABSTRACT

As VR technology advances, the demand for multitasking within virtual environments escalates. Negotiating multiple tasks within the immersive virtual setting presents cognitive challenges, where users experience difficulty executing multiple concurrent tasks. This phenomenon highlights the importance of cognitive functions like attention and working memory, which are vital for navigating intricate virtual environments effectively. In addition to attention and working memory, assessing the extent of physical and mental strain induced by the virtual environment and the concurrent tasks performed by the participant is key. While previous research has focused on investigating factors influencing attention and working memory in virtual reality, more comprehensive approaches addressing the prediction of physical and mental strain alongside these cognitive aspects remain. This gap inspired our investigation, where we utilized an open dataset - VRWalking, which included eye and head tracking and physiological measures like heart rate(HR) and galvanic skin response(GSR). The VRwalking dataset has timestamped labeled data for physical and mental load, working memory, and attention metrics. In our investigation, we employed straightforward deep learning models to predict these labels, achieving noteworthy performance with 91%, 96%, 93%, and 91% accuracy in predicting physical load, mental load, working memory, and attention, respectively. Additionally, we conducted SHAP (SHapley Additive exPlanations) analysis to identify the most critical features driving these predictions. Our findings contribute to understanding the overall cognitive state of a participant and effective data collection practices for future researchers, as well as provide insights for virtual reality developers. Developers can utilize these predictive approaches to adaptively optimize user experience in real-time and minimize cognitive strain, ultimately enhancing the effectiveness and usability of virtual reality applications.

PMID:40063446 | DOI:10.1109/TVCG.2025.3549850

IEEE Trans Pattern Anal Mach Intell. 2025 Mar 10;PP. doi: 10.1109/TPAMI.2025.3549300. Online ahead of print.

ABSTRACT

Despite the great success achieved, deep learning technologies usually suffer from data scarcity issues in real-world applications, where existing methods mainly explore sample relationships in a vanilla way from the perspectives of either the input or the loss function. In this paper, we propose a batch transformer module, BatchFormerV1, to equip deep neural networks themselves with the abilities to explore sample relationships in a learnable way. Basically, the proposed method enables data collaboration, e.g., head-class samples will also contribute to the learning of tail classes. Considering that exploring instance-level relationships has very limited impacts on dense prediction, we generalize and refer to the proposed module as BatchFormerV2, which further enables exploring sample relationships for pixel-/patch-level dense representations. In addition, to address the train-test inconsistency where a mini-batch of data samples are neither necessary nor desirable during inference, we also devise a two-stream training pipeline, i.e., a shared model is first jointly optimized with and without BatchFormerV2 which is then removed during testing. The proposed module is plug-and-play without requiring any extra inference cost. Lastly, we evaluate the proposed method on over ten popular datasets, including 1) different data scarcity settings such as long-tailed recognition, zero-shot learning, domain generalization, and contrastive learning; and 2) different visual recognition tasks ranging from image classification to object detection and panoptic segmentation. Code is available at https://zhihou7.github.io/BatchFormer.

PMID:40063428 | DOI:10.1109/TPAMI.2025.3549300

J Fluoresc. 2025 Mar 10. doi: 10.1007/s10895-025-04229-7. Online ahead of print.

ABSTRACT

Camellia oil (CAO), known for its high nutritional and commercial value, has raised increasing concerns about adulteration. Developing an accurate and non-destructive method to identify CAO adulterants is crucial for safeguarding public health and well-being. This study simulates potential real-world adulteration cases by designing representative adulteration scenarios, followed by the acquisition and analysis of corresponding excitation-emission matrix fluorescence (EEMF) spectra. Parallel factor analysis (PARAFAC) was employed to characterize and explore the variations of fluorophores in the EEMF spectra of different adulterated scenarioss, which showed a linear correlation between the relative concentration of PARAFAC components and adulteration levels. A deep learning model named ResTransformer, which combines residual modules with Transformer, was proposed for both the qualitative detection of adulteration types and the quantitative detection of adulteration concentrations from local and global perspectives. The global ResTransformer qualitative models achieved accuracies of over 96.92% based on EEMF spectra and PARAFAC, and quantitative models showed determination coefficient of validation ([Formula: see text]) > 0.978, root mean square error of validation ([Formula: see text]) < 3.0643%, and the ratio performance deviation (RPD) > 7.6741. Compared to traditional chemometric models, the ResTransformer model demonstrated superior performance. The integration of EEMF and ResTransformer presents a highly promising strategy for rapid and reliable detection of CAO adulteration.

PMID:40063235 | DOI:10.1007/s10895-025-04229-7

EJNMMI Phys. 2025 Mar 10;12(1):21. doi: 10.1186/s40658-025-00728-5.

ABSTRACT

PURPOSE: Various specialized and general collimators are used for myocardial perfusion imaging (MPI) with single-photon emission computed tomography (SPECT) to assess different types of coronary artery disease (CAD). Alongside the wide variability in imaging characteristics, the apriori "learnt" information of left ventricular (LV) shape can affect the final diagnosis of the imaging protocol. This study evaluates the effect of prior information incorporation into the segmentation process, compared to deep learning (DL) approaches, as well as the differences of 4 collimation techniques on 5 different datasets.

METHODS: This study was implemented on 80 patients database. 40 patients were coming from mixed black-box collimators, 10 each, from multi-pinhole (MPH), low energy high resolution (LEHR), CardioC and CardioD collimators. The testing was evaluated on a new continuous graph-based approach, which automatically segments the left ventricular volume with prior information on the cardiac geometry. The technique is based on the continuous max-flow (CMF) min-cut algorithm, which performance was evaluated in precision, recall, IoU and Dice score metrics.

RESULTS: In the testing it was shown that, the developed method showed a good improvement over deep learning reaching higher scores in most of the evaluation metrics. Further investigating the different collimation techniques, the evaluation of receiver operating characterstic (ROC) curves showed different stabilities on the various collimators. Running Wilcoxon signed-rank test on the outlines of the LVs showed differentiability between the collimation procedures. To further investigate these phenomena the model parameters of the LVs were reconstructed and evaluated by the uniform manifold approximation and projection (UMAP) method, which further proved that collimators can be differentiated based on the projected LV shapes alone.

CONCLUSIONS: The results show that prior information incorporation can enhance the performance of segmentation methods and collimation strategies have a high effect on the projected cardiac geometry.

PMID:40063231 | DOI:10.1186/s40658-025-00728-5

J Imaging Inform Med. 2025 Mar 10. doi: 10.1007/s10278-025-01470-1. Online ahead of print.

ABSTRACT

Brain tumor is categorized as one of the most fatal form of cancer due to its location and difficulty in terms of diagnostics. Medical expert relies on two key approaches which include biopsy and MRI. However, these techniques have several setbacks which include the need of medical experts, inaccuracy, miss-diagnosis as a result of anxiety or workload which may lead to patient morbidity and mortality. This opens a gap for the need of precise diagnosis and staging to guide appropriate clinical decisions. In this study, we proposed the application of deep learning (DL)-based techniques for the classification of MRI vs non-MRI and tumor vs no tumor. In order to accurately discriminate between classes, we acquired brain tumor multimodal image (CT and MRI) datasets, which comprises of 9616 MRI and CT scans in which 8000 are selected for discrimination between MRI and non-MRI and 4000 for the discrimination between tumor and no tumor cases. The acquired images undergo image pre-processing, data split, data augmentation and model training. The images are trained using 4 DL networks which include MobileNetV2, ResNet, Ineptionv3 and VGG16. Performance evaluation of the DL architectures and comparative analysis has shown that pre-trained MobileNetV2 achieved the best result across all metrics with 99.94% accuracy for the discrimination between MRI and non-MRI and 99.00% for the discrimination between tumor and no tumor. Moreover, I-BrainNet which is a DL/IoT-based framework is developed for the real-time classification of brain tumor.

PMID:40063173 | DOI:10.1007/s10278-025-01470-1

J Imaging Inform Med. 2025 Mar 10. doi: 10.1007/s10278-025-01469-8. Online ahead of print.

ABSTRACT

CT image denoising is a crucial task in medical imaging systems, aimed at enhancing the quality of acquired visual signals. The emergence of diffusion models in machine learning has revolutionized the generation of high-quality CT images. However, diffusion-based CT image denoising methods suffer from two key shortcomings. First, they do not incorporate image formation priors from CT imaging, which limits their adaptability to the CT image denoising task. Second, they are trained on CT images with varying structures and textures at the signal phase, which hinders the model generalization capability. To address the first limitation, we propose a novel conditioning module for our diffusion model that leverages image formation priors from the sinogram domain to generate rich features. To tackle the second issue, we introduce a two-phase training mechanism in which the network gradually learns different anatomical textures and structures. Extensive experimental results demonstrate the effectiveness of both approaches in enhancing CT image quality, with improvements of up to 17% in PSNR and 38% in SSIM, highlighting their superiority over state-of-the-art methods.

PMID:40063172 | DOI:10.1007/s10278-025-01469-8

Int J Cardiovasc Imaging. 2025 Mar 10. doi: 10.1007/s10554-025-03369-y. Online ahead of print.

ABSTRACT

Invasive coronary physiology is underused and carries risks/costs. Artificial Intelligence (AI) might enable non-invasive physiology from invasive coronary angiography (CAG), possibly outperforming humans, but has seldom been explored, especially for instantaneous wave-free Ratio (iFR). We aimed to develop binary iFR lesion classification AI models and compare them with human performance. single-center retrospective study of patients undergoing CAG and iFR. A validated encoder-decoder convolutional neural network (CNN) performed segmentation. Manual annotation of target vessel and pressure sensor location on a segmented telediastolic frame followed. Three AI models classified lesions as positive (≤ 0.89) or negative (> 0.89). Model 1 uses preprocessed vessel diameters with a transformer. Models 2/3 are EfficientNet-B5 CNNs using concatenated angiography and segmentation - Model 3 employs class-frequency-weighted Cross-Entropy Loss. Previous findings demonstrated Model 3's superiority for left anterior descending (LAD) and Model 1's for circumflex (Cx)/right coronary artery (RCA) - they were therefore unified into a vessel-based model. Ten-fold patient-level cross-validation enabled full sample training/testing. Three experienced operators performed binary iFR classification using single frames of raw/segmented images. Comparison metrics were accuracy, sensitivity, specificity, positive predictive value (PPV) and negative predictive value (NPV). Across 250 measurements, AI accuracy was 72%, PPV 48%, NPV 90%, sensitivity 77%, and specificity 71%. Human accuracy ranged from 54 to 74%. NPV was high for the Cx/RCA (AI: 96/98%; operators: 94/97%), but AI significantly outperformed humans in the LAD (78% vs. 60-64%). An AI model capable of binary iFR lesions classification mildly outperformed interventional cardiologists, supporting further validation studies.

PMID:40063156 | DOI:10.1007/s10554-025-03369-y

Proc Natl Acad Sci U S A. 2025 Mar 18;122(11):e2424470122. doi: 10.1073/pnas.2424470122. Epub 2025 Mar 10.

ABSTRACT

The cystic fibrosis transmembrane conductance regulator (CFTR) is a chloride channel whose dysfunction leads to intracellular accumulation of chloride ions, dehydration of cell surfaces, and subsequent damage to airway and ductal organs. Beyond its function as a chloride channel, interactions between CFTR, epithelium sodium channel, and solute carrier (SLC) transporter family membrane proteins and cytoplasmic proteins, including calmodulin and Na+/H+ exchanger regulatory factor-1 (NHERF-1), coregulate ion homeostasis. CFTR has also been observed to form mesoscale membrane clusters. However, the contributions of multivalent protein and lipid interactions to cluster formation are not well understood. Using a combination of computational modeling and biochemical reconstitution assays, we demonstrate that multivalent interactions with CFTR protein binding partners, calcium, and membrane cholesterol can induce mesoscale CFTR cluster formation on model membranes. Phosphorylation of the intracellular domains of CFTR also promotes mesoscale cluster formation in the absence of calcium, indicating that multiple mechanisms can contribute to CFTR cluster formation. Our findings reveal that coupling of multivalent protein and lipid interactions promotes CFTR cluster formation consistent with membrane-associated biological phase separation.

PMID:40063811 | DOI:10.1073/pnas.2424470122

Microb Genom. 2025 Mar;11(3). doi: 10.1099/mgen.0.001365.

ABSTRACT

Diabetes mellitus is a complex metabolic disorder and one of the fastest-growing global public health concerns. The gut microbiota is implicated in the pathophysiology of various diseases, including diabetes. This study utilized 16S rRNA metagenomic data from a volunteer citizen science initiative to investigate microbial markers associated with diabetes status (positive or negative) and type (type 1 or type 2 diabetes mellitus) using supervised machine learning (ML) models. The diversity of the microbiome varied according to diabetes status and type. Differential microbial signatures between diabetes types and negative group revealed an increased presence of Brucellaceae, Ruminococcaceae, Clostridiaceae, Micrococcaceae, Barnesiellaceae and Fusobacteriaceae in subjects with diabetes type 1, and Veillonellaceae, Streptococcaceae and the order Gammaproteobacteria in subjects with diabetes type 2. The decision tree, elastic net, random forest (RF) and support vector machine with radial kernel ML algorithms were trained to screen and type diabetes based on microbial profiles of 76 subjects with type 1 diabetes, 366 subjects with type 2 diabetes and 250 subjects without diabetes. Using the 1000 most variable features, tree-based models were the highest-performing algorithms. The RF screening models achieved the best performance, with an average area under the receiver operating characteristic curve (AUC) of 0.76, although all models lacked sensitivity. Reducing the dataset to 500 features produced an AUC of 0.77 with sensitivity increasing by 74% from 0.46 to 0.80. Model performance improved for the classification of negative-status and type 2 diabetes. Diabetes type models performed best with 500 features, but the metric performed poorly across all model iterations. ML has the potential to facilitate early diagnosis of diabetes based on microbial profiles of the gut microbiome.

PMID:40063675 | DOI:10.1099/mgen.0.001365

J Vis Exp. 2025 Feb 21;(216). doi: 10.3791/67911.

ABSTRACT

Mitochondrial membrane potential (MMP, ΔΨm) is critical for mitochondrial functions, including ATP synthesis, ion transport, reactive oxygen species (ROS) generation, and the import of proteins encoded by the nucleus. Existing methods for measuring ΔΨm typically use lipophilic cation dyes, such as Rhodamine 800 and tetramethylrhodamine methyl ester (TMRM), but these are limited by low specificity and are not well-suited for in vivo applications. To address these limitations, we have developed a novel protocol utilizing genetically encoded voltage indicators (GEVIs). Genetically encoded voltage indicators (GEVIs), which generate fluorescent signals in response to membrane potential changes, have demonstrated significant potential for monitoring plasma membrane and neuronal potentials. However, their application to mitochondrial membranes remains unexplored. Here, we developed protein-based mitochondrial-targeted GEVIs capable of detecting ΔΨm fluctuations in cells and the motor cortex of living animals. The mitochondrial potential indicator (MPI)offers a non-invasive approach to study ΔΨm dynamics in real-time, providing a method to investigate mitochondrial function under both normal and pathological conditions.

PMID:40063520 | DOI:10.3791/67911

Clin Exp Med. 2025 Mar 10;25(1):79. doi: 10.1007/s10238-025-01577-3.

ABSTRACT

High-altitude pulmonary hypertension (HAPH) is characterized by an increase in pulmonary artery pressure due to prolonged exposure to hypoxic environment at high altitudes. The development of HAPH involves various factors such as pressure changes, inflammation, oxidative stress, gene regulation, and signal transduction. The pathophysiological mechanisms of this condition operate at molecular, cellular, and genetic levels. Diagnosis of HAPH often relies on echocardiography, cardiac catheterization, and other methods to assess pulmonary artery pressure and its impact on cardiac function. Treatment options for HAPH encompass both nondrug and drug therapies. While advancements have been made in understanding the pathological mechanisms through research on animal models and clinical trials, there are still limitations to be addressed. Future research should focus on exploring molecular targets, personalized medicine, long-term management strategies, and interdisciplinary approaches. By leveraging advanced technologies like systems biology, omics technology, big data, and artificial intelligence, a comprehensive analysis of HAPH pathogenesis can lead to the identification of new treatment targets and strategies, ultimately enhancing patient quality of life and prognosis. Furthermore, research on health monitoring and preventive measures for populations living at high altitudes should be intensified to reduce the incidence and mortality of HAPH.

PMID:40063280 | DOI:10.1007/s10238-025-01577-3

In Silico Pharmacol. 2025 Mar 6;13(1):41. doi: 10.1007/s40203-025-00323-7. eCollection 2025.

ABSTRACT

The Marburg virus (MARV), a member of the family Filoviridae, is a highly pathogenic virus causing severe hemorrhagic fever with extremely high mortality in humans and non-human primates. The MARV exhibits clinical and epidemiological features almost identical to those of the Ebola virus, no licensed vaccines or antiviral treatments have been developed yet for MARV. However, only a few treatments that remain uncertain of the disease are available to help bring a case for a new therapeutic approach. Considering the non-availability of any standard drug we have planned to identify potential inhibitors of VP24 (PDB ID: 4OR8) through a computational drug repurposing process. The workflow includes: identifying a druggable pocket on VP24, screening of FDA-approved antivirals via molecular docking, assessing the stability using molecular dynamics simulations, and estimating binding affinity through MM-PBSA calculations. After going through the analysis, the compound Bictegravir manifests as a hit compound which will undergo in vitro and in vivo validation to confirm its efficacy against MARV.

SUPPLEMENTARY INFORMATION: The online version contains supplementary material available at 10.1007/s40203-025-00323-7.

PMID:40061630 | PMC:PMC11885215 | DOI:10.1007/s40203-025-00323-7

bioRxiv [Preprint]. 2025 Feb 28:2025.02.26.640194. doi: 10.1101/2025.02.26.640194.

ABSTRACT

Despite numerous research efforts and several effective vaccines and therapies developed against COronaVIrus Disease 2019 (COVID-19), drug repurposing remains an attractive alternative to identify new treatments for SARS-CoV-2 virus variants and other viral infections that may emerge in the future. Cellular polyamines support viral propagation and tumor growth. Here we tested the antiviral activity of an irreversible inhibitor of polyamine biosynthesis, α-difluoromethylornithine (DFMO) and a non-steroidal anti-inflammatory drug (NSAID) Sulindac, which have been previously evaluated for colon cancer chemoprevention. The drugs were tested as single agents and in combination in human Calu-3 lung adenocarcinoma and Caco-2 colon adenocarcinoma cell lines and the K18-hACE2 transgenic mouse model of severe COVID-19. DFMO/Sulindac combination significantly suppressed SARS-CoV-2 N1 Nucleocapsid mRNA and ACE2 mRNA levels in the infected human cell lines by interacting synergistically when cells were pretreated with drugs and additively when treatment was applied to the infected cells. The antiviral activity of DFMO and Sulindac was tested in vivo as prophylaxis (drug supplementation at the doses equivalent to the human chemoprevention trial started 7 days before infection) or as treatment (drug supplementation started 24 hours post-infection). Prophylaxis with DFMO and Sulindac as single agents significantly increased survival rates in the young male mice (p=0.01, and p=0.027, respectively), and the combination was effective in the aged male mice (p=0.042). Young female mice benefited the most from the prophylaxis with Sulindac alone (p=0.001) and DFMO/Sulindac combination (p=0.018), while aged female mice did not benefit significantly from any interventions. The treatment regime was ineffective in suppressing SARS-CoV-2 infection in K18-hACE2 mice. Overall, animal studies demonstrated the protective age- and sex-dependent antiviral efficacy of DFMO and Sulindac against SARS-CoV-2.

PMID:40060444 | PMC:PMC11888430 | DOI:10.1101/2025.02.26.640194

J Biochem Mol Toxicol. 2025 Mar;39(3):e70204. doi: 10.1002/jbt.70204.

ABSTRACT

Sunitinib (SUN) is a chemotherapeutic agent showing renal toxicity that limits its clinical applications. The present research aimed to clarify the potential ameliorative effects of secukinumab (SEC) and dapagliflozin (DAPA) against SUN-induced renal toxicity and the underpinning molecular mechanisms. For this purpose, adult Wistar albino rats were received SUN (25 mg/kg 3 times/week, po) and co-treated with SEC (3 mg/kg/every 2 weeks, subcutaneously) or DAPA (10 mg/kg/day, po) for 4 weeks and compared with age-matched control group (CON). Markers of kidney functions were assessed in serum samples. Kidneys were harvested for biochemical and histological examination. Compared to CON group, SUN-treated rats displayed signs of kidney dysfunction along with renal histological changes that were ameliorated by SEC or DAPA. Both drugs significantly lowered the renal levels of IL-17, but SEC exerted more inhibitory effect than DAPA. Additionally, SUN-subjected rats showed significant increases in the renal expression of NLRP3 inflammasome and the other inflammatory mediators including IL-1β, END-1, and MCP-1. This was associated with marked decline of the renal levels of beclin-1. Co-treatment with SEC or DAPA significantly suppressed NLRP3-induced inflammation while enhanced beclin-1-mediated autophagy. The modulatory effect of DAPA on NLRP3 and beclin-1 was superior to that of SEC. Moreover, both drugs significantly and similarly attenuated the enhanced cleaved caspase-3 expression and interstitial fibrosis in renal tissue of SUN-subjected rats. Collectively, these findings may repurpose SEC and DAPA as candidate therapies to alleviate the renal toxicity of SUN and to rescue the renal functionality in SUN-treated cancer cases.

PMID:40059817 | DOI:10.1002/jbt.70204

Front Cardiovasc Med. 2025 Feb 21;12:1529114. doi: 10.3389/fcvm.2025.1529114. eCollection 2025.

ABSTRACT

INTRODUCTION: Heart failure (HF) is a highly prevalent disease affecting roughly 7 million Americans. A transcriptome-wide analysis revealed RYR1 upregulation in HF patients with severe pulmonary hypertension. Therefore, we aimed to further characterize the role of RYR1 in HF progression and mortality.

METHODS: In a mouse model of HF, expression of Ryr1 was compared in cardiac pulmonary, and vascular tissue between HF and control mice. Candidate single nucleotide polymorphisms (SNPs) in the RYR1 gene region were identified, including variants affecting RYR1 expression in relevant tissue types. A Cox proportional hazard model was used to analyze genetic associations of candidate SNPs with all-cause mortality in HF patients. An exploratory analysis assessed significantly associated SNPs with risk of HF and arrhythmia development.

RESULTS: In the preclinical HF model, left ventricular expression of Ryr1 was increased compared to control (fold change = 2.08; P = 0.01). In 327 HF patients, decreased mortality risk was associated with two RYR1 SNPs: rs12974674 (HR: 0.59; 95% CI: 0.40-0.87; P = 0.007) and rs2915950 (HR: 0.62, 95% CI: 0.43-0.88; P = 0.008). Based on eQTL data, these SNPs were associated with decreased RYR1 expression in vascular tissue. Two missense variants, in linkage disequilibrium with rs2915950 (rs2915952 and rs2071089) were significantly associated with decreased mortality risk (P = 0.03) and decreased risk of atrial fibrillation/flutter (OR: 0.66, 95% CI: 0.44-0.96; P = 0.03 and OR: 0.67, 95% CI: 0.45-0.98; P = 0.04, respectively). Survival associations with these SNPs were replicated in HF patients self-identifying as Black in the UK Biobank, and the arrhythmia associations were replicated in the overall UK Biobank population.

CONCLUSION: Increased RYR1 expression may contribute to HF progression, potentially through the mechanisms associated with calcium handling and arrhythmia development. Our findings suggest that RYR1 should be further studied as a potential therapeutic target for reducing HF-related mortality.

PMID:40060969 | PMC:PMC11885062 | DOI:10.3389/fcvm.2025.1529114

ACS Omega. 2025 Feb 18;10(8):8619-8629. doi: 10.1021/acsomega.4c11184. eCollection 2025 Mar 4.

ABSTRACT

Hyperlipidemia, a condition characterized by elevated lipid levels, presents significant cardiovascular risks. Syringaldehyde (SA), a phenolic aldehyde derived from plants, exhibits antioxidant, antihyperglycemic, and anti-inflammatory properties. However, its potential toxicity and effects on hyperlipidemia have not been studied. In this study, we evaluated the safety profile and antihyperlipidemic effects of SA. To assess acute toxicity, Sprague-Dawley rats were divided into two groups (n = 5 in each group): the control group received a vehicle, while the treatment group was administered a single oral dose of SA 2000 mg/kg, and rats were observed up to 14 days. To investigate the antihyperlipidemic effects of SA, rats were allocated into six groups (n = 5 in each group). Group 1 (control) received a vehicle, group 2 (hyperlipidemic) was treated with tyloxapol (i.p 400 mg/kg), while groups 3-6 received atorvastatin 10 mg/kg and SA 10, 20, and 40 mg/kg, respectively, post tyloxapol injection. The acute toxicity results showed that SA exhibits LD50 above 2000 mg/kg. Hematological analyses showed no significant changes, except for a notable increase in the platelet count. Additionally, SA significantly decreases cholesterol, triglyceride, and creatinine levels, along with elevated alanine transaminase, alkaline phosphatase, and urea levels. Markers of oxidative stress confirmed SA's antioxidant properties, and histopathological examination revealed normal cellular structure of selected organs. In the hyperlipidemic model, SA effectively and dose dependently reduced hyperlipidemia by lowering total cholesterol, triglycerides, and LDL levels and improved hepatocellular structure affected by tyloxapol. Moreover, gene expression analysis demonstrated significant downregulation in SREBP-2 gene expression along with reduced HMG-CoA reductase activity. Overall, this study supports the safety and low toxicity of SA and its promising antihyperlipidemic effects.

PMID:40060812 | PMC:PMC11886728 | DOI:10.1021/acsomega.4c11184

bioRxiv [Preprint]. 2025 Mar 3:2025.02.27.640661. doi: 10.1101/2025.02.27.640661.

ABSTRACT

OBJECTIVE: The fast accumulation of vast pharmacogenomics data of cancer cell lines provide unprecedented opportunities for drug sensitivity prediction (DSP), a crucial prerequisite for the advancement of precision oncology. Recently, Generative Large Language Models (LLM) have demonstrated performance and generalization prowess across diverse tasks in the field of natural language processing (NLP). However, the structured format of the pharmacogenomics data poses challenge for the utility of LLM in DSP. Therefore, the objective of this study is multi-fold: to adapt prompt engineering for structured pharmacogenomics data toward optimizing LLM's DSP performance, to evaluate LLM's generalization in real-world DSP scenarios, and to compare LLM's DSP performance against that of state-of-the-science baselines.

METHODS: We systematically investigated the capability of the Generative Pre-trained Transformer (GPT) as a DSP model on four publicly available benchmark pharmacogenomics datasets, which are stratified by five cancer tissue types of cell lines and encompass both oncology and non-oncology drugs. Essentially, the predictive landscape of GPT is assessed for effectiveness on the DSP task via four learning paradigms: zero-shot learning, few-shot learning, fine-tuning and clustering pretrained embeddings. To facilitate GPT in seamlessly processing the structured pharmacogenomics data, domain-specific novel prompt engineering is employed by implementing three prompt templates (i.e., Instruction, Instruction-Prefix, Cloze) and integrating pharmacogenomics-related features into the prompt. We validated GPT's performance in diverse real-world DSP scenarios: cross-tissue generalization, blind tests, and analyses of drug-pathway associations and top sensitive/resistant cell lines. Furthermore, we conducted a comparative evaluation of GPT against multiple Transformer-based pretrained models and existing DSP baselines.

RESULTS: Extensive experiments on the pharmacogenomics datasets across the five tissue cohorts demonstrate that fine-tuning GPT yields the best DSP performance (28% F1 increase, p-value= 0.0003) followed by clustering pretrained GPT embeddings (26% F1 increase, p-value= 0.0005), outperforming GPT in-context learning (i.e., few-shot). However, GPT in the zero-shot setting had a big F1 gap, resulting in the worst performance. Within the scope of prompt engineering, performance enhancement was achieved by directly instructing GPT about the DSP task and resorting to a concise context format (i.e., instruction-prefix), leading to F1 performance gain of 22% (p-value=0.02); while incorporation of drug-cell line prompt context derived from genomics and/or molecular features further boosted F1 score by 2%. Compared to state-of-the-science DSP baselines, GPT significantly asserted superior mean F1 performance (16% gain, p-value<0.05) on the GDSC dataset. In the crosstissue analysis, GPT showcased comparable generalizability to the within-tissue performances on the GDSC and PRISM datasets, while statistically significant F1 performance improvements on the CCLE (8%, p-value=0.001) and DrugComb (19%, p-value=0.009) datasets. Evaluation on the challenging blind tests suggests GPT's competitiveness on the CCLE and DrugComb datasets compared to random splitting. Furthermore, analyses of the drug-pathway associations and log probabilities provided valuable insights that align with previous DSP findings.

CONCLUSION: The diverse experiment setups and in-depth analysis underscore the importance of generative LLM, such as GPT, as a viable in silico approach to guide precision oncology.

AVAILABILITY: https://github.com/bioIKEA/SensitiveCancerGPT.

PMID:40060567 | PMC:PMC11888479 | DOI:10.1101/2025.02.27.640661

Infect Immun. 2025 Mar 10:e0059124. doi: 10.1128/iai.00591-24. Online ahead of print.

ABSTRACT

Respiratory infections with multiresistant Pseudomonas aeruginosa are a major clinical problem, affecting mainly patients with pre-existing lung diseases such as cystic fibrosis (CF) or chronic obstructive pulmonary disease but also immunocompromised or elderly patients. We have previously shown that sphingosine, which is abundantly present on epithelial cells of the respiratory tract in healthy humans and wild-type mice, but almost undetectable on the surface of epithelial cells of the respiratory tract from CF patients and CF mice, efficiently kills many bacterial species in vitro and in vivo. Here, we show that sphingosine very rapidly induces marked changes in the membrane of P. aeruginosa with a rolling of the membrane followed by destruction of the bacteria. Sphingosine induced a degradation of cardiolipin via the maintenance of lipid asymmetry (Mla) system in P. aeruginosa. Degradation of cardiolipin induced by sphingosine is prevented in P. aeruginosa mutants of MlaY and reduced in mutants of MlaZ and MlaA. Mutants of MlaY and MlaZ were resistant to sphingosine-induced death of P. aeruginosa. In summary, our data indicate that sphingosine induces the death of P. aeruginosa by a persisting degradation of cardiolipin by the Mla system leading to severe membrane changes in bacteria, while leaving mammalian cells, devoid of cardiolipin in their plasma membrane, alive.

PMID:40062881 | DOI:10.1128/iai.00591-24

Pediatr Pulmonol. 2025 Mar;60(3):e71037. doi: 10.1002/ppul.71037.

ABSTRACT

BACKGROUND: Although modulator therapies have proven effective in cystic fibrosis (CF) access is limited due to reimbursement issues in Turkey. We aimed to examine anxiety and depression levels of people with CF (pwCF) and their caregivers according to their access to modulator treatment.

METHODS: Participants genetically eligible for elexacaftor/tezacaftor/ivacaftor (ETI) were divided into Group 1 (access via court decision, not yet on treatment) and Group 2 (unable to access due to reimbursement issues). Genetically ineligible participants formed Group 3. All pwCF and parents of those under 18 were screened for depression by the Patient Health Questionnaire-9 (PHQ-9) and for anxiety by the Generalized Anxiety Disorder-7 (GAD-7). Surveys for Group 1 patients were conducted just before starting ETI. Binary logistic regression analysis was performed to evaluate the effects of independent variables on anxiety and depression in pwCF and their primary caregivers.

RESULTS: A total of 389 pwCF and 285 caregivers were included. Group 3 (ineligible) had the highest depression rate (72.9%, n = 35), while Group 1 (pre-ETI) had the lowest (50.0%, n = 35). Median PHQ-9 scores were significantly lower in Group 1 (p < 0.006). Anxiety rates were higher in Groups 2 and 3 compared to Group 1 (p = 0.011 and p = 0.003, respectively). Access to ETI reduced the odds of anxiety by 67.7% (p = 0.029). Caregiver GAD-7 scores showed a weak negative correlation with pwCF age (r = -0.117).

CONCLUSION: Limited access to modulator therapies is associated with higher depression and anxiety symptoms among pwCF. Addressing these barriers is critical to improving their well-being.

PMID:40062574 | DOI:10.1002/ppul.71037