Diabetes. 2025 Jan 24:db240497. doi: 10.2337/db24-0497. Online ahead of print.

ABSTRACT

PPARγ is the pharmacological target of thiazolidinediones (TZDs), potent insulin sensitizers that prevent metabolic disease morbidity but are accompanied by side effects such as weight gain, in part due to non-physiological transcriptional agonism. Using high throughput genome engineering, we targeted nonsense mutations to every exon of PPARG, finding an ATG in Exon 2 (chr3:12381414, CCDS2609 c.A403) that functions as an alternative translational start site. This downstream translation initiation site gives rise to a PPARγ protein isoform (M135), preferentially generated from alleles containing nonsense mutations upstream of c.A403. PPARγ M135 retains the DNA and ligand binding domains of full-length PPARγ but lacks the N-terminal AF-1 domain. Despite being truncated, PPARγ M135 shows increased transactivation of target genes, but only in the presence of agonists. Accordingly, human missense mutations disrupting AF-1 domain function actually increase agonist-induced cellular PPARγ activity compared to wild-type (WT), and carriers of these AF-1 disrupting variants are protected from metabolic syndrome. Thus, we propose the existence of PPARγ M135 as a fully functional, alternatively translated isoform that may be therapeutically generated to treat insulin resistance-related disorders.

PMID:39854214 | DOI:10.2337/db24-0497

STAR Protoc. 2025 Jan 23;6(1):103584. doi: 10.1016/j.xpro.2024.103584. Online ahead of print.

ABSTRACT

Here, we present a protocol to generate dual-target compounds (DT-CPDs) interacting with two distinct target proteins using a transformer-based chemical language model. We describe steps for installing software, preparing data, and pre-training the model on pairs of single-target compounds (ST-CPDs), which bind to an individual protein, and DT-CPDs. We then detail procedures for assembling ST- and corresponding DT-CPD data for specific protein pairs and evaluating the model's performance on hold-out test sets. For complete details on the use and execution of this protocol, please refer to Srinivasan and Bajorath.1.

PMID:39854202 | DOI:10.1016/j.xpro.2024.103584

Adv Sci (Weinh). 2025 Jan 24:e2413367. doi: 10.1002/advs.202413367. Online ahead of print.

ABSTRACT

Glioblastoma multiforme (GBM) is a highly aggressive and malignant brain tumor originating from glial cells, characterized by high recurrence rates and poor patient prognosis. The heterogeneity and complex biology of GBM, coupled with the protective nature of the blood-brain barrier (BBB), significantly limit the efficacy of traditional therapies. The rapid development of nanoenzyme technology presents a promising therapeutic paradigm for the rational and targeted treatment of GBM. In this review, the underlying mechanisms of GBM pathogenesis are comprehensively discussed, emphasizing the impact of the BBB on treatment strategies. Recent advances in nanoenzyme-based approaches for GBM therapy are explored, highlighting how these nanoenzymes enhance various treatment modalities through their multifunctional capabilities and potential for precise drug delivery. Finally, the challenges and therapeutic prospects of translating nanoenzymes from laboratory research to clinical application, including issues of stability, targeting efficiency, safety, and regulatory hurdles are critically analyzed. By providing a thorough understanding of both the opportunities and obstacles associated with nanoenzyme-based therapies, future research directions are aimed to be informed and contribute to the development of more effective treatments for GBM.

PMID:39854126 | DOI:10.1002/advs.202413367

Am J Hum Biol. 2025 Jan;37(1):e24212. doi: 10.1002/ajhb.24212.

ABSTRACT

INTRODUCTION: The maximal fat oxidation (MFO) and the exercise intensity that provokes MFO (FATMAX) are inversely associated with cardiometabolic risk factors in healthy young sedentary adults. However, how both cardiorespiratory fitness (CRF) level and sex influence MFO during exercise and the FATMAX is seldom analyzed.

OBJECTIVES: This study is aimed at determining the influence of CRF and sex on MFO.

METHODS: Twenty healthy young adults (i.e., 12 men and 8 women) completed a graded treadmill protocol to determine MFO, MFO relative to lean mass (MFOlean), FATMAX and maximum oxygen uptake (VO2max).

RESULTS: The k-means cluster analysis was used to divide the sample into two different groups for CRF level (56.54 ± 2.54 and 46.94 ± 3.07 mL/kg/min, p < 0.001, respectively). The high-level group revealed higher MFO relative to lean mass (MFOlean) (3.34 ± 1.44 and 2.73 ± 0.87 g · min-1 · kg, p = 0.001, respectively), and FATMAX in km · h-1 (FATMAXv) (7.67 ± 0.90 and 7.00 ± 0.97 km · h-1, p = 0.044, respectively) but not for MFO (0.67 ± 0.19 and 0.71 ± 0.20 p = 0.124, respectively). When divided for sex, men exhibited higher values for MFO (0.76 ± 0.21 vs. 0.69 ± 0.19 g · min-1, p = 0.039) and FATMAXv (7.67 ± 0.96 vs. 7.30 ± 0.98 km · h-1, p = 0.036), while women showed higher values for MFOlean (3.92 ± 1.35 vs. 2.40 ± 0.46 g · min-1 · kg, p = 0.015).

CONCLUSION: This study highlights the significant influence of CRF level and sex on MFO and FATMAX, offering valuable insights for tailoring exercise programs and optimizing health and performance interventions.

PMID:39853816 | DOI:10.1002/ajhb.24212

Ann Neurol. 2025 Jan 24. doi: 10.1002/ana.27087. Online ahead of print.

ABSTRACT

Congenital titinopathy has recently emerged as one of the most common congenital muscle disorders.

OBJECTIVE: To better understand the presentation and clinical needs of the under-characterized extreme end of the congenital titinopathy severity spectrum.

METHODS: We comprehensively analyzed the clinical, imaging, pathology, autopsy, and genetic findings in 15 severely affected individuals from 11 families.

RESULTS: Prenatal features included hypokinesia or akinesia and growth restriction. Six pregnancies were terminated. Nine infants were born at or near term with severe-to-profound weakness and required resuscitation. Seven died following withdrawal of life support. Two surviving children require ongoing respiratory support. Most cohort members had at least 1 disease-causing variant predicted to result in some near-normal-length titin expression. The exceptions, from 2 unrelated families, had homozygous truncating variants predicted to induce complete nonsense mediated decay. However, subsequent analyses suggested that the causative variant in each family had an additional previously unrecognized impact on splicing likely to result in some near-normal-length titin expression. This impact was confirmed by minigene assay for 1 variant.

INTERPRETATION: This study confirms the clinical variability of congenital titinopathy. Severely affected individuals succumb prenatally/during infancy, whereas others survive into adulthood. It is likely that this variability is because of differences in the amount and/or length of expressed titin. If confirmed, analysis of titin expression could facilitate clinical prediction and increasing expression might be an effective treatment strategy. Our findings also further-support the hypothesis that some near-normal-length titin expression is essential to early prenatal survival. Sometimes expression of normal/near-normal-length titin is due to disease-causing variants having an additional impact on splicing. ANN NEUROL 2025.

PMID:39853809 | DOI:10.1002/ana.27087

World J Biol Psychiatry. 2025 Jan 24:1-16. doi: 10.1080/15622975.2025.2453198. Online ahead of print.

ABSTRACT

BACKGROUND: Genes associated with global developmental delay (GDD) and intellectual disability (ID) are increasingly being identified through next-generation sequencing (NGS) technologies. This study aimed to identify novel mutations in GDD/ID phenotypes through whole-exome sequencing (WES) and additional in silico analyses.

MATERIAL AND METHODS: WES was performed on 27 subjects, among whom 18 were screened for potential novel mutations. In silico analyses included protein-protein interactions (PPIs), gene-miRNA interactions (GMIs), and enrichment analyses. The identified novel variants were further modelled using I-Tasser-MTD and SWISS-MODEL, with structural superimposition performed.

RESULTS: Novel mutations were detected in 18 patients, with 10 variants reported for the first time. Among these, three were classified as pathogenic (DNMT1:c.856dup, KCNQ2:c.1635_1636insT, and TMEM94:c.2598_2599insC), and six were likely pathogenic. DNMT1 and MRE11 were highlighted as key players in PPIs and GMIs. GMIs analysis emphasised the roles of hsa-miR-30a-5p and hsa-miR-185-5p. The top-scoring pathways included the neuronal system (R-HSA-112316, p = 7.73E-04) and negative regulation of the smooth muscle cell apoptotic process (p = 3.37E-06). Homology modelling and superimposition revealed a significant functional loss in the mutated DNMT1 enzyme structure.

CONCLUSION: This study identified 10 novel pathogenic/likely pathogenic variants associated with GDD/ID, supported by clinical findings and in silico analyses focused on DNMT1 mutations.

PMID:39853208 | DOI:10.1080/15622975.2025.2453198

Vaccines (Basel). 2025 Jan 1;13(1):33. doi: 10.3390/vaccines13010033.

ABSTRACT

Adjuvants are a diverse group of substances that can be added to vaccines to enhance antigen-specific immune responses and improve vaccine efficacy. The first adjuvants, discovered almost a century ago, were soluble crystals of aluminium salts. Over the following decades, oil emulsions, vesicles, oligodeoxynucleotides, viral capsids, and other complex organic structures have been shown to have adjuvant potential. However, the detailed mechanisms of how adjuvants enhance immune responses remain poorly understood and may be a barrier that reduces the rational selection of vaccine components. Previous studies on mechanisms of action of adjuvants have focused on how they activate innate immune responses, including the regulation of cell recruitment and activation, cytokine/chemokine production, and the regulation of some "immune" genes. This approach provides a narrow perspective on the complex events involved in how adjuvants modulate antigen-specific immune responses. A comprehensive and efficient way to investigate the molecular mechanism of action for adjuvants is to utilize systems biology approaches such as transcriptomics in so-called "systems vaccinology" analysis. While other molecular biology methods can verify if one or few genes are differentially regulated in response to vaccination, systems vaccinology provides a more comprehensive picture by simultaneously identifying the hundreds or thousands of genes that interact with complex networks in response to a vaccine. Transcriptomics tools such as RNA sequencing (RNA-Seq) allow us to simultaneously quantify the expression of practically all expressed genes, making it possible to make inferences that are only possible when considering the system as a whole. Here, we review some of the challenges in adjuvant studies, such as predicting adjuvant activity and toxicity when administered alone or in combination with antigens, or classifying adjuvants in groups with similar properties, while underscoring the significance of transcriptomics in systems vaccinology approaches to propel vaccine development forward.

PMID:39852812 | DOI:10.3390/vaccines13010033

Cells. 2025 Jan 15;14(2):119. doi: 10.3390/cells14020119.

ABSTRACT

As of 2023, there were 39.9 million people living with Human Immunodeficiency Virus type 1 (HIV-1). Although great strides have been made in treatment options for HIV-1, and our understanding of the HIV-1 life cycle has vastly improved since the start of this global health crisis, a functional cure remains elusive. One of the main barriers to a cure is latency, which allows the virus to persist despite combined antiretroviral therapy (cART). Recently, we have found that exosomes, which are small, membrane-enclosed particles released by virtually all cell types and known to mediate intercellular communication, caused an increase in RNA Polymerase II loading onto the HIV-1 promoter. This resulted in the production of both short- and long-length viral transcripts in infected cells under cART. This current study examines the effects of exosome-associated kinases on bystander cells. The phospho-kinase profiling of exosomes revealed differences in the kinase payload of exosomes derived from uninfected and HIV-1-infected cells, with CDK10, GSK3β, and MAPK8 having the largest concentration differences. These kinases were shown to be biologically active and capable of phosphorylating substrates, and they modulated changes in the cell cycle dynamics of exposed cells. Given the relevance of such effects for the immune response, our results implicate exosome-associated kinases as new possible key contributors to HIV-1 pathogenesis that affect bystander cells. These findings may guide new therapeutic avenues to improve the current antiretroviral treatment regimens.

PMID:39851547 | DOI:10.3390/cells14020119

Bioengineering (Basel). 2024 Dec 25;12(1):7. doi: 10.3390/bioengineering12010007.

ABSTRACT

In spite of significant advancements in diagnosis and treatment, cancer remains one of the major threats to human health due to its ability to cause disease with high morbidity and mortality. A multifactorial and multitargeted approach is required towards intervention of the multitude of signaling pathways associated with carcinogenesis inclusive of angiogenesis and metastasis. In this context, plants provide an immense source of phytotherapeutics that show great promise as anticancer drugs. There is increasing epidemiological data indicating that diets rich in vegetables and fruits could decrease the risks of certain cancers. Several studies have proved that natural plant polyphenols, such as flavonoids, lignans, phenolic acids, alkaloids, phenylpropanoids, isoprenoids, terpenes, and stilbenes, could be used in anticancer prophylaxis and therapeutics by recruitment of mechanisms inclusive of antioxidant and anti-inflammatory activities and modulation of several molecular events associated with carcinogenesis. The current review discusses the anticancer activities of principal phytochemicals with focus on signaling circuits towards targeted cancer prophylaxis and therapy. Also addressed are plant-derived anti-cancer vaccines, nanoparticles, monoclonal antibodies, and immunotherapies. This review article brings to light the importance of plants and plant-based platforms as invaluable, low-cost sources of anti-cancer molecules of particular applicability in resource-poor developing countries.

PMID:39851281 | DOI:10.3390/bioengineering12010007

Elife. 2025 Jan 24;12:RP88149. doi: 10.7554/eLife.88149.

ABSTRACT

Apart from ancestry, personal or environmental covariates may contribute to differences in polygenic score (PGS) performance. We analyzed the effects of covariate stratification and interaction on body mass index (BMI) PGS (PGSBMI) across four cohorts of European (N = 491,111) and African (N = 21,612) ancestry. Stratifying on binary covariates and quintiles for continuous covariates, 18/62 covariates had significant and replicable R2 differences among strata. Covariates with the largest differences included age, sex, blood lipids, physical activity, and alcohol consumption, with R2 being nearly double between best- and worst-performing quintiles for certain covariates. Twenty-eight covariates had significant PGSBMI-covariate interaction effects, modifying PGSBMI effects by nearly 20% per standard deviation change. We observed overlap between covariates that had significant R2 differences among strata and interaction effects - across all covariates, their main effects on BMI were correlated with their maximum R2 differences and interaction effects (0.56 and 0.58, respectively), suggesting high-PGSBMI individuals have highest R2 and increase in PGS effect. Using quantile regression, we show the effect of PGSBMI increases as BMI itself increases, and that these differences in effects are directly related to differences in R2 when stratifying by different covariates. Given significant and replicable evidence for context-specific PGSBMI performance and effects, we investigated ways to increase model performance taking into account nonlinear effects. Machine learning models (neural networks) increased relative model R2 (mean 23%) across datasets. Finally, creating PGSBMI directly from GxAge genome-wide association studies effects increased relative R2 by 7.8%. These results demonstrate that certain covariates, especially those most associated with BMI, significantly affect both PGSBMI performance and effects across diverse cohorts and ancestries, and we provide avenues to improve model performance that consider these effects.

PMID:39851248 | DOI:10.7554/eLife.88149

Front Pharmacol. 2025 Jan 9;15:1444311. doi: 10.3389/fphar.2024.1444311. eCollection 2024.

ABSTRACT

OBJECTIVE: The expanding field of hematopoietic cell transplantation (HCT) for non-malignant diseases, including those amenable to gene therapy or gene editing, faces challenges due to limited donor availability and the toxicity associated with cell collection methods. Umbilical cord blood (CB) represents a readily accessible source of hematopoietic stem and progenitor cells (HSPCs); however, the cell dose obtainable from a single cord blood unit is frequently insufficient. This limitation can be addressed by enhancing the potency of HSPCs, specifically their capacity to reconstitute hematopoiesis. In our study, we investigated the combined effects of treprostinil, a prostaglandin analog, and cinacalcet, a calcium-sensing receptor modulator, on the reconstitution of hematopoiesis.

METHODS: A Lineage Cell Depletion Kit was employed to isolate lineage-negative (lin-) HSPCs from mouse bone marrow. A Human CB CD34 Positive Selection Kit was utilized to isolate CD34+ cells from the CB of healthy donors. In vitro, the effects of treprostinil, cinacalcet, and their combination on the migration, adhesion, and differentiation of HSPCs were assessed. In vivo, homing and engraftment were examined. Eight-week-old female and male C57BL/6J, BALB/c, or female NSG mice served as recipient models.

RESULTS: When administered concomitantly, treprostinil and cinacalcet exhibited mutual antagonism: the survival of recipient animals was lower when both drugs were administered together compared to either agent alone. Conversely, a sequential regimen involving priming with treprostinil/forskolin followed by cinacalcet treatment in vivo enhanced survival, irrespective of whether hematopoiesis was reconstituted by human or murine HSPCs. In vitro assays demonstrated enhanced migration and adhesion in response to the presence of treprostinil and cinacalcet, suggesting potential synergistic effects. Colony formation confirmed synergism.

CONCLUSION: Augmenting the bone marrow reconstitution potential of HSPCs with treprostinil and cinacalcet shows promise for rescuing patients undergoing HCT. This approach is particularly beneficial for those patients at high risk of transplant failure due to limited numbers of available HSPCs. Furthermore, enhancing the potency of HSPCs has the potential to alleviate the burden and risks associated with HSPC donation, as it would reduce the number of cells needed for collection.

PMID:39850556 | PMC:PMC11755040 | DOI:10.3389/fphar.2024.1444311

J Biomol Struct Dyn. 2025 Jan 24:1-15. doi: 10.1080/07391102.2024.2335282. Online ahead of print.

ABSTRACT

The recent spread of SARS-CoV-2 has led to serious concerns about newly emerging infectious coronaviruses. Drug repurposing is a practical method for rapid development of antiviral agents. The viral spike protein of SARS-CoV-2 binds to its major receptor ACE2 to promote membrane fusion. Following the entry process, the spike protein is further activated by cellular proteases such as TMPRSS2 and Furin to promote viral entry into human cells. A crucial factor in preventing SARS-CoV-2 from entering target cells using HIV-1 fusion inhibitors is the similarity between the fusion mechanisms of SARS-CoV-2 and HIV-1. In this investigation, the HIV-1 fusion inhibitors CMK, Luteolin, and Naphthofluorescein were selected to understand the molecular mode of interactions and binding energy of Furin with these experimental inhibitors. The binding affinity of the three inhibitors with Furin was verified by molecular docking studies. The docking scores of CMK, Luteolin and Naphthofluorescein are -7.4 kcal/mol, -9.3 kcal/mol, and -10.7 kcal/mol, respectively. Therefore, these compounds were subjected to MD, drug-likeness, ADMET, and MM-PBSA analysis. According to the results of a 200 ns MD simulation, all tested compounds show stability with the complex and can be employed as promising inhibitors targeting SARS-CoV-2 Furin protease. In addition, pharmacokinetic analysis revealed that these compounds possess favorable drug-likeness properties. Thus, this study of Furin inhibitors helps in the evaluation of these compounds for use as novel drugs against SARS-CoV-2.

PMID:39849987 | DOI:10.1080/07391102.2024.2335282

Front Pharmacol. 2025 Jan 9;15:1397995. doi: 10.3389/fphar.2024.1397995. eCollection 2024.

ABSTRACT

INTRODUCTION: TNFα inhibitor (TNFi) immunogenicity in rheumatoid arthritis (RA) is a major obstacle to its therapeutic effectiveness. Although methotrexate (MTX) can mitigate TNFi immunogenicity, its adverse effects necessitate alternative strategies. Targeting nuclear factor of activated T cells (NFAT) transcription factors may protect against biologic immunogenicity. Therefore, developing a potent NFAT inhibitor to suppress this immunogenicity may offer an alternative to MTX.

METHODS: We performed a structure-based virtual screen of the NFATC2 crystal structure to identify potential small molecules that could interact with NFATC2. For validation, we investigated the effect of the identified compound on NFAT transcriptional activity, nuclear localization, and binding to the NFAT consensus sequence. In vivo studies assessed the ability of the compound to protect against TNFi immunogenicity, while ex vivo studies evaluated its effect on CD4+ T cell proliferation and B cell antibody secretion.

RESULTS: We identified duvelisib (DV) as a novel NFATC2 and NFATC1 inhibitor that attenuates NFAT transcriptional activity without inhibiting calcineurin or NFAT nuclear localization. Our results suggest that DV inhibits NFAT independently of PI3K by interfering with nuclear NFAT binding to the NFAT consensus promoter sequence. DV significantly protected mice from adalimumab immunogenicity and attenuated ex vivo CD4+ T cell proliferation and B cell antibody secretion.

DISCUSSION: DV is a promising NFAT inhibitor that can protect against TNFi immunogenicity without inhibiting calcineurin phosphatase activity. Our results suggest that the future development of DV analogs may be of interest as agents to attenuate unwanted immune responses.

PMID:39850568 | PMC:PMC11754251 | DOI:10.3389/fphar.2024.1397995

Front Genet. 2025 Jan 9;15:1523071. doi: 10.3389/fgene.2024.1523071. eCollection 2024.

NO ABSTRACT

PMID:39850490 | PMC:PMC11754187 | DOI:10.3389/fgene.2024.1523071

BMC Health Serv Res. 2025 Jan 23;25(1):133. doi: 10.1186/s12913-025-12292-w.

ABSTRACT

BACKGROUND: Remote patient monitoring implies continuous follow-up of health-related parameters of patients outside healthcare facilities. Patients share health-related data with their healthcare unit and obtain feedback (which may be automatically generated if data are within a predefined range). The goals of remote patient monitoring are improvements for patients and reduced healthcare costs. The aim of this paper is to provide an overview of systematic reviews regarding remote patient monitoring for selected patient groups currently considered for the introduction of remote patient monitoring in Region Västra Götaland, Sweden. The selected sixteen patient groups were: patients with asthma, chronic obstructive pulmonary disease, children and adolescents with complex needs, children and adolescents with cystic fibrosis, children and adolescents with periodic fever, elderly patients with multiple diseases, patients with eye diseases, heart failure, haematological disease, hypertension, inflammatory bowel disease, neurorehabilitation, Parkinson's disease, psoriasis, sleep apnea, and specialist maternity care. Outcomes considered in this overview were patient-relevant clinical benefits as well as risks.

METHODS: A literature search for systematic reviews of clinical trials on remote patient monitoring in the selected patient groups was conducted by two information specialists, followed by assessment of relevance by a team of clinical and methodological experts in Region Västra Götaland, Sweden. The methodological rigour of identified systematic reviews was assessed using QUICKSTAR - a tool for stepwise appraisal of systematic reviews. In a QUICKSTAR assessment, a level of at least five is considered a prerequisite for reliable conclusions regarding the question at issue.

RESULTS: The literature search resulted in 4,049 hits, of which 84 SRs were considered relevant for the question at issue. A QUICKSTAR level of at least five was reached by 13 (15%) of the relevant systematic reviews. Some patient benefit of remote patient monitoring was reported for five patient groups (asthma, chronic obstructive lung disease, heart failure, hypertension, and elderly patients with multiple diseases). For four patient groups (children with complex needs, children with cystic fibrosis, specialist maternity care, and sleep apnea), systematic reviews of adequate quality concluded that scientific evidence on clinical patient benefits of remote monitoring is very limited. For seven patient groups, no systematic reviews of sufficient quality were identified.

CONCLUSION: Clinical benefits and risks of remote patient monitoring as a replacement for, or in addition to, standard of care compared to standard of care (face-to-face visits) are poorly studied for most of the selected patient groups based on systematic reviews of acceptable quality. Patient-relevant clinical benefits are limited or impossible to evaluate for most diagnoses based on currently available scientific information. Possible clinical risks and costs are poorly studied.

PMID:39849519 | DOI:10.1186/s12913-025-12292-w

Brief Bioinform. 2024 Nov 22;26(1):bbaf023. doi: 10.1093/bib/bbaf023.

ABSTRACT

Antibody generation requires the use of one or more time-consuming methods, namely animal immunization, and in vitro display technologies. However, the recent availability of large amounts of antibody sequence and structural data in the public domain along with the advent of generative deep learning algorithms raises the possibility of computationally generating novel antibody sequences with desirable developability attributes. Here, we describe a deep learning model for computationally generating libraries of highly human antibody variable regions whose intrinsic physicochemical properties resemble those of the variable regions of the marketed antibody-based biotherapeutics (medicine-likeness). We generated 100000 variable region sequences of antigen-agnostic human antibodies belonging to the IGHV3-IGKV1 germline pair using a training dataset of 31416 human antibodies that satisfied our computational developability criteria. The in-silico generated antibodies recapitulate intrinsic sequence, structural, and physicochemical properties of the training antibodies, and compare favorably with the experimentally measured biophysical attributes of 100 variable regions of marketed and clinical stage antibody-based biotherapeutics. A sample of 51 highly diverse in-silico generated antibodies with >90th percentile medicine-likeness and > 90% humanness was evaluated by two independent experimental laboratories. Our data show the in-silico generated sequences exhibit high expression, monomer content, and thermal stability along with low hydrophobicity, self-association, and non-specific binding when produced as full-length monoclonal antibodies. The ability to computationally generate developable human antibody libraries is a first step towards enabling in-silico discovery of antibody-based biotherapeutics. These findings are expected to accelerate in-silico discovery of antibody-based biotherapeutics and expand the druggable antigen space to include targets refractory to conventional antibody discovery methods requiring in vitro antigen production.

PMID:39851074 | DOI:10.1093/bib/bbaf023

Food Chem X. 2024 Nov 13;24:101981. doi: 10.1016/j.fochx.2024.101981. eCollection 2024 Dec 30.

ABSTRACT

With the rising demand of saffron, it is essential to standardize the confirmation of its origin and identify any adulteration to maintain a good quality led market product. However, a rapid and reliable strategy for identifying the adulteration saffron is still lacks. Herein, a combination of headspace-gas chromatography-ion mobility spectrometry (HS-GC-IMS) and convolutional neural network (CNN) was developed. Sixty-nine volatile compounds (VOCs) including 7 groups of isomers were detected rapidly and directly. A CNN prediction model based on GC-IMS data was proposed. With the merit of minimal data prepossessing and automatic feature extraction capability, GC-IMS images were directly input to the CNN model. The origin prediction results were output with the average accuracy about 90 %, which was higher than traditional methods like PCA (61 %) and SVM (71 %). This established CNN also showed ability in identifying counterfeit saffron with a high accuracy of 98 %, which can be used to authenticate saffron.

PMID:39850938 | PMC:PMC11754009 | DOI:10.1016/j.fochx.2024.101981

Front Public Health. 2025 Jan 15;12:1426168. doi: 10.3389/fpubh.2024.1426168. eCollection 2024.

ABSTRACT

INTRODUCTION: The growing demand for real-time, affordable, and accessible healthcare has underscored the need for advanced technologies that can provide timely health monitoring. One such area is predicting arterial blood pressure (BP) using non-invasive methods, which is crucial for managing cardiovascular diseases. This research aims to address the limitations of current healthcare systems, particularly in remote areas, by leveraging deep learning techniques in Smart Health Monitoring (SHM).

METHODS: This paper introduces a novel neural network architecture, ResNet-LSTM, to predict BP from physiological signals such as electrocardiogram (ECG) and photoplethysmogram (PPG). The combination of ResNet's feature extraction capabilities and LSTM's sequential data processing offers improved prediction accuracy. Comprehensive error analysis was conducted, and the model was validated using Leave-One-Out (LOO) cross-validation and an additional dataset.

RESULTS: The ResNet-LSTM model showed superior performance, particularly with PPG data, achieving a mean absolute error (MAE) of 6.2 mmHg and a root mean square error (RMSE) of 8.9 mmHg for BP prediction. Despite the higher computational cost (~4,375 FLOPs), the improved accuracy and generalization across datasets demonstrate the model's robustness and suitability for continuous BP monitoring.

DISCUSSION: The results confirm the potential of integrating ResNet-LSTM into SHM for accurate and non-invasive BP prediction. This approach also highlights the need for accurate anomaly detection in continuous monitoring systems, especially for wearable devices. Future work will focus on enhancing cloud-based infrastructures for real-time analysis and refining anomaly detection models to improve patient outcomes.

PMID:39850864 | PMC:PMC11755415 | DOI:10.3389/fpubh.2024.1426168

Front Artif Intell. 2025 Jan 9;7:1498956. doi: 10.3389/frai.2024.1498956. eCollection 2024.

ABSTRACT

INTRODUCTION: Active learning can significantly decrease the labeling cost of deep learning workflows by prioritizing the limited labeling budget to high-impact data points that have the highest positive impact on model accuracy. Active learning is especially useful for semantic segmentation tasks where we can selectively label only a few high-impact regions within these high-impact images. Most established regional active learning algorithms deploy a static-budget querying strategy where a fixed percentage of regions are queried in each image. A static budget could result in over- or under-labeling images as the number of high-impact regions in each image can vary.

METHODS: In this paper, we present a novel dynamic-budget superpixel querying strategy that can query the optimal numbers of high-uncertainty superpixels in an image to improve the querying efficiency of regional active learning algorithms designed for semantic segmentation.

RESULTS: For two distinct datasets, we show that by allowing a dynamic budget for each image, the active learning algorithm is more effective compared to static-budget querying at the same low total labeling budget. We investigate both low- and high-budget scenarios and the impact of superpixel size on our dynamic active learning scheme. In a low-budget scenario, our dynamic-budget querying outperforms static-budget querying by 5.6% mIoU on a specialized agriculture field image dataset and 2.4% mIoU on Cityscapes.

DISCUSSION: The presented dynamic-budget querying strategy is simple, effective, and can be easily adapted to other regional active learning algorithms to further improve the data efficiency of semantic segmentation tasks.

PMID:39850848 | PMC:PMC11754207 | DOI:10.3389/frai.2024.1498956

Int J Burns Trauma. 2024 Dec 15;14(6):125-132. doi: 10.62347/VCJP9652. eCollection 2024.

ABSTRACT

PURPOSE: To evaluate the identification of nasal bone fractures and their clinical diagnostic significance for three-dimensional (3D) reconstruction of maxillofacial computed tomography (CT) images by applying artificial intelligence (AI) with deep learning (DL).

METHODS: CT maxillofacial 3D reconstruction images of 39 patients with normal nasal bone and 43 patients with nasal bone fracture were retrospectively analysed, and a total of 247 images were obtained in three directions: the orthostatic, left lateral and right lateral positions. The CT scan images of all patients were reviewed by two senior specialists to confirm the presence or absence of nasal fractures. Binary classification prediction was performed using the YOLOX detection model + GhostNetv2 classification model with a DL algorithm. Accuracy, sensitivity, and specificity were used to evaluate the efficacy of the AI model. Manual independent review, and AI model-assisted manual independent review were used to identify nasal fractures.

RESULTS: Compared with those of manual independent detection, the accuracy, sensitivity, and specificity of AI-assisted film reading improved between junior and senior physicians. The differences were statistically significant (P<0.05), and all were higher than manual independent detection.

CONCLUSIONS: Based on deep learning methods, an artificial intelligence model can be used to assist in the diagnosis of nasal bone fractures, which helps to promote the practical clinical application of deep learning methods.

PMID:39850782 | PMC:PMC11751554 | DOI:10.62347/VCJP9652