Nat Commun. 2025 May 7;16(1):4260. doi: 10.1038/s41467-025-59525-4.

ABSTRACT

Blood cell phenotypes are routinely tested in healthcare to inform clinical decisions. Genetic variants influencing mean blood cell phenotypes have been used to understand disease aetiology and improve prediction; however, additional information may be captured by genetic effects on observed variance. Here, we mapped variance quantitative trait loci (vQTL), i.e. genetic loci associated with trait variance, for 29 blood cell phenotypes from the UK Biobank (N ~ 408,111). We discovered 176 independent blood cell vQTLs, of which 147 were not found by additive QTL mapping. vQTLs displayed on average 1.8-fold stronger negative selection than additive QTL, highlighting that selection acts to reduce extreme blood cell phenotypes. Variance polygenic scores (vPGSs) were constructed to stratify individuals in the INTERVAL cohort (N ~ 40,466), where the genetically most variable individuals had increased conventional PGS accuracy (by ~19%) relative to the genetically least variable individuals. Genetic prediction of blood cell traits improved by ~10% on average combining PGS with vPGS. Using Mendelian randomisation and vPGS association analyses, we found that alcohol consumption significantly increased blood cell trait variances highlighting the utility of blood cell vQTLs and vPGSs to provide novel insight into phenotype aetiology as well as improve prediction.

PMID:40335489 | DOI:10.1038/s41467-025-59525-4

Trends Biotechnol. 2025 May 6:S0167-7799(25)00134-9. doi: 10.1016/j.tibtech.2025.04.007. Online ahead of print.

ABSTRACT

Given the crucial role of water-soluble vitamins in the human body and the rising demand for natural sources, their biosynthesis has gained the attention of researchers. This review offers a comprehensive look at recent progress in water-soluble vitamin biosynthesis, emphasizing synthetic biotechnology for green biomanufacturing. Specifically, it encompasses the optimization of biological components, pathways, and systems, as well as energy metabolism regulation, stress-tolerance enhancement, high-throughput screening, and the upscaling of production processes. It also envisages intelligent biomanufacturing platforms, highlighting the role of systems biology and artificial intelligence (AI), and proposes future research directions, such as integrating AI-driven metabolic models, enzyme engineering, and cell-free systems, to address limitations in the efficiency, toxicity, and scalability of water-soluble vitamin production.

PMID:40335344 | DOI:10.1016/j.tibtech.2025.04.007

Int J Biol Macromol. 2025 May 5:143949. doi: 10.1016/j.ijbiomac.2025.143949. Online ahead of print.

ABSTRACT

Porous starch (PS) has been utilized as an oral protective carrier to enhance the oxidative stability of liposoluble nutrients. However, PS releases more glucose during digestion, thereby increasing the risk of chronic diseases. Chlorogenic acid (CA) has excellent antioxidant properties and enhances the starch digestion resistance. To simultaneously enhance the oxidative protection and anti-digestibility, PS was blended with CA. Morphological analysis revealed that PSs with pores absorbed liposoluble substances. Surface area, total pore volume, and oxidative stability analyses demonstrated that rice starch (RS) enzymatically hydrolyzed for 12 h (PS12) loaded more substances and exerted a better protective effect in cooperation with CA. Simulated digestion confirmed that PS12-CA1 had the best anti-digestibility among PS12-CAs and a similar digestibility as RS. Additionally, CA treatment resulted in more anti-digestive V-type crystals in PSs, which resisted digestion. This study showed that the combination of PS and CA simultaneously enhanced oxidative protection and reduced the digestibility of PS. Thus, CA treatment makes PS a better oral nutrient delivery.

PMID:40334898 | DOI:10.1016/j.ijbiomac.2025.143949

Bioresour Technol. 2025 May 5:132620. doi: 10.1016/j.biortech.2025.132620. Online ahead of print.

ABSTRACT

Fermented food remains poorly understood, largely due to the lack of knowledge about microbes in food fermentation. Here, this study constructed Moutai Fermented Grain Catalog (MTFGC), a representative liquor fermented by one of the most complex fermentations. MTFGC comprised 8,379,551 non-redundant genes and 5,159 metagenome-assembled genomes, with 20% species and 20% genes being novel. Additionally, 25,625 biosynthetic gene clusters (BGCs) and 28 BGC-enriched species were identified. Moreover, the microbial community assembly was deterministic, with significant species and gene changes in early fermentation stages, while stabilizing in later stages. Further BGC-knockout experiments verified Bacillus licheniformis, a BGC-enriched species, employed its BGCs for synthesizing the aroma-related lipopeptide lichenysin. This study has established the largest genetic resource for fermented food, uncovering its uniqueness and high metabolic potential. These findings facilitate the transition potential from traditional fermentation to precision-driven synthetic biology in food systems.

PMID:40334798 | DOI:10.1016/j.biortech.2025.132620

J Lipid Res. 2025 May 5:100820. doi: 10.1016/j.jlr.2025.100820. Online ahead of print.

ABSTRACT

BACKGROUND: Elevated levels of lipoprotein (a) [Lp(a)], an apolipoprotein B particle, are causally linked to atherosclerotic cardiovascular disease (ASCVD). Lp(a) is thought to promote ASCVD through multiple mechanisms, including its effects on cholesterol transport, inflammation, and thrombosis.

OBJECTIVE: Define the mechanisms that integrate Lp(a)-mediated cholesterol accumulation, inflammation, and thrombosis.

METHODS: In this study, we employed systems biology approaches, including proteomics, transcriptomics, and mass cytometry, to define the immune cellular and molecular phenotypes in ASCVD subjects with high and low Lp(a) levels and the molecular mechanisms through which Lp(a) mediates monocyte-driven inflammation and thrombosis.

RESULTS: In 64 stable ASCVD subjects (41 with high Lp(a) [median Lp(a) 228.7 nmol/L] and 23 with low Lp(a) [median Lp(a) 17.8 nmol/L]), we found that circulating markers of inflammation (CCL28, IL-17D) and vascular dysfunction (tissue factor [TF]; 6.4 vs 5.7 normalized protein expression (NPX); p=0.01) were elevated in subjects with high Lp(a) levels compared with those with low Lp(a) levels. Although total monocyte and hsCRP levels were similar between the groups, CD14+ monocytes from ASCVD subjects with an elevated Lp(a) were primed and expressed more TF at baseline and in response to stress. Mechanistically, we found that Lp(a) itself can activate monocytes through Toll-like receptor 2 (TLR2) and nuclear factor kappa B (NFκB) signaling, driving both the induction of TF and TF activity.

CONCLUSIONS: Overall, these studies are the first to link Lp(a) to monocyte-mediated inflammation and thrombosis. They demonstrate a novel mechanism through TLR2, NFκB, and monocyte TF by which Lp(a) amplifies immunothrombotic risk.

PMID:40334781 | DOI:10.1016/j.jlr.2025.100820

Metab Eng. 2025 May 5:S1096-7176(25)00075-8. doi: 10.1016/j.ymben.2025.04.011. Online ahead of print.

ABSTRACT

Genetically engineered microbes are increasingly utilized to produce a broad range of high-value compounds. However, most studies start with only a very narrow group of genetically tractable type strains that have not been selected for maximum titers or industrial robustness. In this study, we used high-throughput screening and parallel metabolic engineering to identify and optimize Saccharomyces cerevisiae chassis strains for the production of limonene, a monoterpene with applications in flavors, fragrances, and biofuels. We screened 921 genetically and phenotypically distinct S. cerevisiae strains for limonene tolerance and lipid content to identify optimal chassis strains for precision fermentation of limonene. In parallel, we also evaluated 16 different plant limonene synthases. Our results revealed that two of the selected strains showed approximately a 2-fold increase in titers compared to CEN.PK2-1C, the type strain that is often used as a chassis for limonene production, with the same genetic modifications in the mevalonate pathway. Intriguingly, the most effective engineering strategy proved strain-specific. Metabolic profiling revealed that this difference is likely explained by differences in native mevalonate production. Ultimately, by using strain-specific engineering strategies, we achieved 844 mg/L in a new strain, 40% higher than the titer (605 mg/L) achieved by CEN.PK2-1C. Our findings demonstrate the potential of leveraging genetic diversity in S. cerevisiae for monoterpene bioproduction and highlight the necessity for tailoring metabolic engineering strategies to specific strains.

PMID:40334774 | DOI:10.1016/j.ymben.2025.04.011

PLoS One. 2025 May 7;20(5):e0322546. doi: 10.1371/journal.pone.0322546. eCollection 2025.

ABSTRACT

Proteins are fundamental biomolecules composed of one or more chains of amino acids. They are essential for all living organisms, contributing to various biological functions and regulatory processes. Alterations in protein structures and functions are closely linked to diseases, emphasizing the need for in-depth study. A thorough understanding of these associations is crucial for developing targeted and more effective therapeutic strategies.Computational analyses of biomedical data facilitate the identification of specific patterns in proteins associated with diseases, providing novel insights into their biological roles. This study introduces a computational approach designed to detect relevant sequence patterns within proteins. These patterns, characterized by specific amino acid arrangements, can be critical for protein functionality. The proposed methodology was applied to proteins targeted by drugs used in lung cancer treatment, a disease that remains the leading cause of cancer-related mortality worldwide. Given that non-small cell lung cancer represents 85-90% of all lung cancer cases, it was selected as the primary focus of this study.Significant sequence patterns were identified, establishing connections between drug-target proteins and proteins associated with lung cancer. Based on these findings, a novel computational framework was developed to extend this pattern-based analysis to proteins linked to other diseases. By employing this approach, relationships between lung cancer drug-target proteins and proteins associated with four additional cancer types were uncovered. These associations, characterized by shared amino acid sequence features, suggest potential opportunities for drug repurposing. Furthermore, validation through an extensive literature review confirmed biological links between lung cancer drug-target proteins and proteins related to other malignancies, reinforcing the potential of this methodology for identifying new therapeutic applications.

PMID:40334012 | DOI:10.1371/journal.pone.0322546

Molecules. 2025 Apr 13;30(8):1736. doi: 10.3390/molecules30081736.

ABSTRACT

Alzheimer's disease (AD) affects an increasing number of people as the human population ages. The main pathological feature of AD, amyloid plaques, consists of the key protein amyloid-β and other co-deposited proteins. These co-deposited proteins and their protein interactors could hold some additional functional insights into AD pathophysiology. For this work, proteins found on amyloid plaques were collected from the AmyCo database. A protein-protein and protein-drug interaction network was constructed with data from the IntAct and DrugBank databases, respectively. In total, there were 12 proteins co-deposited on amyloid plaques that reportedly interact with 513 other proteins and are targets of 72 drugs. These drugs were shown to be almost entirely distinct from the panel of drugs currently approved by the FDA for AD and their corresponding protein targets. In conclusion, this work demonstrates the potential for drug repurposing of drugs that target proteins found in amyloid plaques.

PMID:40333680 | DOI:10.3390/molecules30081736

Talanta. 2025 May 5;294:128278. doi: 10.1016/j.talanta.2025.128278. Online ahead of print.

ABSTRACT

The early diagnosis of diseases largely relies on the monitoring and accurate detection of biomarkers within biological systems. The quantification of chloride ions (Cl-) and glutathione (GSH) can effectively assess the progression of diseases such as cystic fibrosis and cancer, as well as the alterations in the body's internal environment. However, developing reliable sensing platforms with high sensitivity and selectivity poses significant challenges. Based on the dual-functional silver-based metal-organic frameworks (Ag MOF), an electrochemical/electrochemiluminescent (EC/ECL) dual-channel nanoplatform was developed for the detection of Cl- and GSH, aided by graphitic carbon nitride (g-C3N4). In the EC mode, the interaction between Ag MOF and Cl- leads to the formation of silver chloride (AgCl), which is characterized by an increased peak current of AgCl solid-state electrochemistry as Cl- concentration rises. The further introduction of GSH generates a non-electroactive complex through competition with Cl-, resulting in a decrease in the peak current of AgCl. In the ECL mode, the quenching of ECL signals from g-C3N4 by Ag MOF is alleviated by Cl-, due to the etching of the Ag-MOF. The ECL recovery effect is further enhanced with the addition of GSH. For Cl-, both EC and ECL responses exhibit good linear relationships with concentrations ranging from 0.5 to 10 mM, with detection limit (LOD) of 0.4 mM and 0.1 mM, respectively. For GSH, EC and ECL also show good linear relationships in range of 0.01-100 μM, with LOD of 9.8 nM and 1.02 nM. The unique properties of Ag MOF, acting both as an electrochemical sensing component that generates sensitive current outputs for Cl- and GSH, and as a quencher for the ECL of g-C3N4, facilitate the sequential detection of Cl- and GSH, providing mutual validation that significantly enhances accuracy and reliability. The specific interactions of Ag MOF with these analytes offer the innovative platform good selectivity, demonstrating significant potential for advancements in biological analysis and disease diagnosis.

PMID:40334507 | DOI:10.1016/j.talanta.2025.128278

PLoS One. 2025 May 7;20(5):e0311711. doi: 10.1371/journal.pone.0311711. eCollection 2025.

ABSTRACT

Arsenic is associated with lung disease and experimental models suggest that arsenic-induced degradation of the chloride channel CFTR (cystic fibrosis transmembrane conductance regulator) is a mechanism of arsenic toxicity. We examined associations between arsenic exposure, sweat chloride concentration (measure of CFTR function), and pulmonary function among 269 adults in Bangladesh. Participants with sweat chloride ≥ 60 mmol/L had higher arsenic exposures than those with sweat chloride < 60 mmol/L (water: median 77.5 µg/L versus 34.0 µg/L, p = 0.025; toenails: median 4.8 µg/g versus 3.7 µg/g, p = 0.024). In linear regression models, a one-unit µg/g increment in toenail arsenic was associated with a 0.59 mmol/L higher sweat chloride concentration, p < 0.001. Among the entire study population, after adjusting for covariates including age, sex, smoking, education, and height, toenail arsenic concentration was associated with increased odds of airway obstruction (OR: 1.97, 95%: 1.06, 3.67, p = 0.03); however, sweat chloride concentration did not mediate this association. Our findings suggest that sweat chloride concentration may serve as novel biomarker for arsenic exposure, warranting further investigation in diverse populations, and that arsenic likely acts on the lung through mechanisms other than inducing CFTR dysfunction. Alternative mechanisms by which environmental arsenic exposure may lead to obstructive lung disease, such as arsenic-induced direct lung injury and/or increase lung proteinase activity, require additional exploration in future work.

PMID:40333927 | DOI:10.1371/journal.pone.0311711

Pathogens. 2025 Apr 15;14(4):387. doi: 10.3390/pathogens14040387.

ABSTRACT

This study analyzed eleven isolates of colistin-resistant Pseudomonas aeruginosa, originating from Portugal and Taiwan, which are associated with various pathologies. The results revealed significant genetic diversity among the isolates, with each exhibiting a distinct genetic profile. A prevalence of sequence type ST235 was observed, characterizing it as a high-risk clone, and serotyping indicated a predominance of type O11, associated with chronic respiratory infections in cystic fibrosis (CF) patients. The phylogenetic analysis demonstrated genetic diversity among the isolates, with distinct clades and complex evolutionary relationships. Additionally, transposable elements such as Tn3 and IS6 were identified in all isolates, highlighting their importance in the mobility of antibiotic resistance genes. An analysis of antimicrobial resistance profiles revealed pan-drug resistance in all isolates, with a high prevalence of genes conferring resistance to β-lactams and aminoglycosides. Furthermore, additional analyses revealed mutations in regulatory networks and specific loci previously implicated in colistin resistance, such as pmrA, cprS, phoO, and others, suggesting a possible contribution to the observed resistant phenotype. This study has a strong impact because it not only reveals the genetic diversity and resistance mechanisms in P. aeruginosa but also identifies mutations in regulatory genes associated with colistin resistance.

PMID:40333140 | DOI:10.3390/pathogens14040387

Int J Pediatr Otorhinolaryngol. 2025 May 4;194:112369. doi: 10.1016/j.ijporl.2025.112369. Online ahead of print.

ABSTRACT

BACKGROUND: Artificial Intelligence (AI) and machine learning (ML) have transformative potential in enhancing diagnostics, treatment planning, and patient management. However, their application in pediatric otolaryngology remains limited as the unique physiological and developmental characteristics of children require tailored AI applications, highlighting a gap in knowledge.

PURPOSE: To provide a narrative review of current literature on the application of AI in pediatric otolaryngology, highlighting knowledge gaps, associated challenges and future directions.

RESULTS: ML models have demonstrated efficacy in diagnosing conditions such as otitis media, adenoid hypertrophy, and pediatric obstructive sleep apnea through deep learning-based image analysis and predictive modeling. AI systems also show potential in surgical settings such as landmark identification during otologic surgery and prediction of middle ear effusion during tympanostomy tube placement. Telemedicine solutions and large language models have shown potential to improve accessibility to care and patient education. The principal challenges include flawed generalization of adult training data and the relative lack of pediatric data.

CONCLUSIONS: AI holds significant promise in pediatric otolaryngology. However, its widespread clinical integration requires addressing algorithmic bias, enhancing model interpretability, and ensuring robust validation across pediatric population. Future research should prioritize federated learning, developmental trajectory modeling, and psychosocial integration to create patient-centered solutions.

PMID:40334638 | DOI:10.1016/j.ijporl.2025.112369

Artif Intell Med. 2025 Apr 29;166:103116. doi: 10.1016/j.artmed.2025.103116. Online ahead of print.

ABSTRACT

Multiple Sclerosis (MS) is a chronic neuroinflammatory disease of the Central Nervous System (CNS) in which the body's immune system attacks and destroys the myelin sheath that protects nerve fibers, leading to a wide range of debilitating symptoms and causing disruption of axonal signal transmission. Accurate prediction, diagnosis, monitoring and treatment (PDMT) of MS are essential to improve patient outcomes. Recent advances in neuroimaging technologies, particularly electroencephalography (EEG), combined with machine learning (ML) techniques - including Deep Learning (DL) models - offer promising avenues for enhancing MS management. This systematic review synthesizes existing research on the application of ML and DL models to EEG data for MS. It explores the methodologies used, with a focus on DL architectures such as Convolutional Neural Networks (CNNs) and hybrid models, and highlights recent advancements in ML techniques and EEG technologies that have significantly improved MS diagnosis and monitoring. The review addresses the challenges and potential biases in using ML-based EEG analysis for MS. Strategies to mitigate these challenges, including advanced preprocessing techniques, diverse training datasets, cross-validation methods, and explainable Artificial Intelligence (AI), are discussed. Finally, the paper outlines potential future applications and trends in ML for MS management. This review underscores the transformative potential of ML-enhanced EEG analysis in improving MS management, providing insights into future research directions to overcome existing limitations and further improve clinical practice.

PMID:40334524 | DOI:10.1016/j.artmed.2025.103116

Food Chem. 2025 Apr 28;486:144311. doi: 10.1016/j.foodchem.2025.144311. Online ahead of print.

ABSTRACT

Resistant starch (RS) is a vital dietary component with notable health benefits, but tradition quantification methods are labor-intensive, costly, and unsuitable for large-scale applications. This study introduced an innovative data-driven framework integrating Near-Infrared (NIR) spectroscopy with Convolutional Neural Networks (CNN) and data augmentation to achieve rapid, cost-effective RS prediction. Achieving exceptional accuracy (Rp2 = 0.992), the CNN model outperformed traditional methods like Partial Least Squares Regression (PLSR) and Support Vector Machine Regression (SVMR). To overcome the "black-box" limitation of deep learning, SHapley Additive exPlanations (SHAP) were innovatively employed, pinpointing critical wavelengths (2000-2500 nm), significantly narrowing the spectral range while providing meaningful insights into the contribution of specific wavelengths to RS prediction. This optimized spectral enhanced data acquisition efficiency, reduces analytical costs, and simplifies operational complexity, establishing a practical and scalable solution for deploying NIR spectroscopy in food quality assessment and production-line applications.

PMID:40334489 | DOI:10.1016/j.foodchem.2025.144311

Comput Methods Programs Biomed. 2025 May 6;267:108830. doi: 10.1016/j.cmpb.2025.108830. Online ahead of print.

ABSTRACT

OBJECTIVES: Achieving efficient analysis of multiple pathological indicators has great significance for breast cancer prognosis and therapeutic decision-making. In this study, we aim to explore a deep multi-task learning (MTL) framework for collaborative prediction of histological grade and proliferation marker (Ki-67) status in breast cancer using multi-phase dynamic contrast-enhanced magnetic resonance imaging (DCE-MRI).

METHODS: In the novel design of hybrid multi-task architecture (HMT-Net), co-representative features are explicitly distilled using a feature extraction backbone. A customized prediction network is then introduced to perform soft-parameter sharing between two correlated tasks. Specifically, task-common and task-specific knowledge is transmitted into tower layers for informative interactions. Furthermore, low-level feature maps containing tumor edges and texture details are recaptured by a hard-parameter sharing branch, which are then incorporated into the tower layer for each subtask. Finally, the probabilities of two histological indicators, predicted in the multi-phase DCE-MRI, are separately fused using a decision-level fusion strategy.

RESULTS: Experimental results demonstrate that the proposed HMT-Net achieves optimal discriminative performance over other recent MTL architectures and deep models based on single image series, with the area under the receiver operating characteristic curve of 0.908 for tumor grade and 0.694 for Ki-67 status.

CONCLUSIONS: Benefiting from the innovative HMT-Net, our proposed method elucidates its strong robustness and flexibility in the collaborative prediction task of breast biomarkers. Multi-phase DCE-MRI is expected to contribute valuable dynamic information for breast cancer pathological assessment in a non-invasive manner.

PMID:40334302 | DOI:10.1016/j.cmpb.2025.108830

Comput Biol Med. 2025 May 6;192(Pt B):110242. doi: 10.1016/j.compbiomed.2025.110242. Online ahead of print.

ABSTRACT

Brain tumours continue to be a primary cause of worldwide death, highlighting the critical need for effective and accurate diagnostic tools. This article presents MK-YOLOv8, an innovative lightweight deep learning framework developed for the real-time detection and categorization of brain tumours from MRI images. Based on the YOLOv8 architecture, the proposed model incorporates Ghost Convolution, the C3Ghost module, and the SPPELAN module to improve feature extraction and substantially decrease computational complexity. An x-small object detection layer has been added, supporting precise detection of small and x-small tumours, which is crucial for early diagnosis. Trained on the Figshare Brain Tumour (FBT) dataset comprising (3,064) MRI images, MK-YOLOv8 achieved a mean Average Precision (mAP) of 99.1% at IoU (0.50) and 88.4% at IoU (0.50-0.95), outperforming YOLOv8 (98% and 78.8%, respectively). Glioma recall improved by 26%, underscoring the enhanced sensitivity to challenging tumour types. With a computational footprint of only 96.9 GFLOPs (representing 37.5% of YOYOLOv8x'sFLOPs) and utilizing 12.6 million parameters, a mere 18.5% of YOYOLOv8's parameters, MK-YOLOv8 delivers high efficiency with reduced resource demands. Also, it trained on the Br35H dataset (801 images) to guarantee the model's robustness and generalization; it achieved a mAP of 98.6% at IoU (0.50). The suggested model operates at 62 frames per second (FPS) and is suited for real-time clinical processes. These developments establish MK-YOLOv8 as an innovative framework, overcoming challenges in tiny tumour identification and providing a generalizable, adaptable, and precise detection approach for brain tumour diagnostics in clinical settings.

PMID:40334297 | DOI:10.1016/j.compbiomed.2025.110242

PLoS One. 2025 May 7;20(5):e0322540. doi: 10.1371/journal.pone.0322540. eCollection 2025.

ABSTRACT

Knee osteoarthritis (KOA) is a leading cause of disability globally. Early and accurate diagnosis is paramount in preventing its progression and improving patients' quality of life. However, the inconsistency in radiologists' expertise and the onset of visual fatigue during prolonged image analysis often compromise diagnostic accuracy, highlighting the need for automated diagnostic solutions. In this study, we present an advanced deep learning model, OA-HybridCNN (OHC), which integrates ResNet and DenseNet architectures. This integration effectively addresses the gradient vanishing issue in DenseNet and augments prediction accuracy. To evaluate its performance, we conducted a thorough comparison with other deep learning models using five-fold cross-validation and external tests. The OHC model outperformed its counterparts across all performance metrics. In external testing, OHC exhibited an accuracy of 91.77%, precision of 92.34%, and recall of 91.36%. During the five-fold cross-validation, its average AUC and ACC were 86.34% and 87.42%, respectively. Deep learning, particularly exemplified by the OHC model, has greatly improved the efficiency and accuracy of KOA imaging diagnosis. The adoption of such technologies not only alleviates the burden on radiologists but also significantly enhances diagnostic precision.

PMID:40334259 | DOI:10.1371/journal.pone.0322540

PLoS One. 2025 May 7;20(5):e0322978. doi: 10.1371/journal.pone.0322978. eCollection 2025.

ABSTRACT

BACKGROUND: Predicting protein-DNA binding sites in vivo is a challenging but urgent task in many fields such as drug design and development. Most promoters contain many transcription factor (TF) binding sites, yet only a few have been identified through time-consuming biochemical experiments. To address this challenge, numerous computational approaches have been proposed to predict TF binding sites from DNA sequences. However, current deep learning methods often face issues such as gradient vanishing as the model depth increases, leading to suboptimal feature extraction.

RESULTS: We propose a model called CBR-KAN (where C represents Convolutional Neural Network (CNN), B represents Bidirectional Long Short Term Memory (BiLSTM), and R represents Residual Mechanism) to predict transcription factor binding sites. Specifically, we designed a multi-scale convolution module (ConvBlock1, 2, 3) combined with BiLSTM network, introduced KAN network to replace traditional multilayer perceptron, and promoted model optimization through residual connections. Testing on 50 common ChIP seq benchmark datasets shows that CBR-KAN outperforms other state-of-the-art methods such as DeepBind, DanQ, DeepD2V, and DeepSEA in predicting TF binding sites.

CONCLUSIONS: The CBR-KAN model significantly improves prediction accuracy for transcription factor binding sites by effectively integrating multiple neural network architectures and mechanisms. This approach not only enhances feature extraction but also stabilizes training and boosts generalization capabilities. The promising results on multiple key performance indicators demonstrate the potential of CBR-KAN in bioinformatics applications.

PMID:40334196 | DOI:10.1371/journal.pone.0322978

PLoS One. 2025 May 7;20(5):e0319476. doi: 10.1371/journal.pone.0319476. eCollection 2025.

ABSTRACT

In the context of consumer economics, it is imperative to consider the functionality of sports venues based on customer demand. However, traditional survey methods are time-consuming, resource-intensive, and coverage-limited. This paper conducted sentiment mining based on Internet big data, deep learning, topic analysis, and social network analysis to capture the satisfaction of consumers and its influencing factors. Findings indicate that activity, courses, and facilities are core factors driving positive comments. Coaches, environment, and activities are key determinants influencing neutral evaluations. Attitude, integrity, and qualifications can trigger negative reviews. The findings offer insights into developing consumer-friendly service for sports venues.

PMID:40333946 | DOI:10.1371/journal.pone.0319476

PLoS One. 2025 May 7;20(5):e0322936. doi: 10.1371/journal.pone.0322936. eCollection 2025.

ABSTRACT

Identifying relevant plant parts is one of the most significant tasks in the pharmaceutical industry. Correct identification minimizes the risk of mis-identification, which might have unfavorable effects, and it ensures that plants are used medicinally. Traditional methods for plant part identification are often time-consuming and require specific expertise. This study proposed a Depth-wise Separable Convolutional Neural Network (DWS-CNN) to enhance the accuracy of medicinal plant part identification. Furthermore, we incorporated the tuned pre-trained models such as VGG16, Res Net-50, and Inception V3 which are designed by Standard convolutional neural network (S-CNN) for comparative purposes. We trained variants of the Standard convolutional neural network (S-CNN) model with high-resolution images of medicinal plant leaves which contains 15,100 leaf images. The study used supervised learning by which leaf images are used as an identity for the other parts of the plants. We used transfer learning to tune training and model parameters. Experimental results showed that our DWS-CNN model achieved better performance compared to S-CNN models, with an accuracy of 99.84% for training data, 99.44% for F1-score and 99.44% for testing data, which improves in both accuracy and training speed. The presence of depth-wise separable convolution and batch normalization at the fully connected layer of the model made the model achieved a good classification performance.

PMID:40333881 | DOI:10.1371/journal.pone.0322936