Search

Video procedural captioning aims to generate detailed descriptive captions for all steps in a long instructional video. The peculiarity of this problem is the procedural dependency between the events to generate consistent captions among the video. However, existing video (dense) captioning methods only consider intra-event or sequential inter-event context and are hard to model the non-sequential context dependency between events. In this paper, inspired by the recent success of graph neural networks in capturing the relations for structured data, we propose a novel M ultimodal G raph N eural N etwork (MGNN) for dense video procedural captioning in capturing the procedural structure between events. Specifically, we construct temporal sequential graph and semantic non-sequential graph for a multimodal heterogeneous graph. Moreover, we adopt the graph neural network to enhance the visual and text features, and fuse both features for further caption generation. Extensive experiments demonstrate the proposed MGNN is effective in generating coherent captions on both the Youcook2 and Activitynet Captions benchmark. [ABSTRACT FROM AUTHOR]

Global sensitivity analysis (GSA) of distribution system with respect to stochastic PV and load variations plays an important role in designing optimal voltage control schemes. This paper proposes a data-driven framework for GSA of distribution system. In particular, two representative surrogate modeling-based approaches are developed, including the traditional Gaussian process-based and the analysis of variance (ANOVA) kernel ones. The key idea is to develop a surrogate model that captures the hidden global relationship between voltage and real and reactive power injections from the historical data. With the surrogate model, the Sobol indices can be conveniently calculated through either the sampling-based method or the analytical method to assess the global sensitivity of voltage to variations of load and PV power injections. The sampling-based method estimates the Sobol indices using Monte Carlo simulations while the analytical method calculates them by resorting to the ANOVA expansion framework. Comparison results with other model-based GSA methods on the unbalanced three-phase IEEE 37-bus and 123-bus distribution systems show that the proposed framework can achieve much higher computational efficiency with negligible loss of accuracy. The results on a real 240-bus distribution system using actual smart meter data further validate the feasibility and scalability of the proposed framework. [ABSTRACT FROM AUTHOR]

Background and Objective: Chronic kidney disease (CKD) is a major public health issue, and accurate prediction of the progression of kidney failure is critical for clinical decision-making and helps improve patient outcomes. As such, we aimed to develop and externally validate a machine-learned model to predict the progression of CKD using common laboratory variables, demographic characteristics, and an electronic health records database., Methods: We developed a predictive model using longitudinal clinical data from a single center for Chinese CKD patients. The cohort included 987 patients who were followed up for more than 24 months. Fifty-three laboratory features were considered for inclusion in the model. The primary outcome in our study was an estimated glomerular filtration rate ≤15 mL/min/1.73 m 2 or kidney failure. Machine learning algorithms were applied to the modeling dataset ( n = 296), and an external dataset ( n = 71) was used for model validation. We assessed model discrimination via area under the curve (AUC) values, accuracy, sensitivity, specificity, positive predictive value, negative predictive value, and F1 score., Results: Over a median follow-up period of 3.75 years, 148 patients experienced kidney failure. The optimal model was based on stacking different classifier algorithms with six laboratory features, including 24-h urine protein, potassium, glucose, urea, prealbumin and total protein. The model had considerable predictive power, with AUC values of 0.896 and 0.771 in the validation and external datasets, respectively. This model also accurately predicted the progression of renal function in patients over different follow-up periods after their initial assessment., Conclusions: A prediction model that leverages routinely collected laboratory features in the Chinese population can accurately identify patients with CKD at high risk of progressing to kidney failure. An online version of the model can be easily and quickly applied in clinical management and treatment., Competing Interests: The authors declare that they have no competing interests., (© 2024 Du et al.)

Ehrlichiosis is a potentially life-threatening disease caused by infection with the obligatory intracellular bacteria Ehrlichia species. Ehrlichia japonica infection of mice provides an animal model of ehrlichiosis as it recapitulates full-spectrum and lethal ehrlichiosis in humans. The E. japonica transposon mutant of EHF0962 , which encodes a previously uncharacterized hypothetical protein, is attenuated in both infection and virulence in mice. EHF0962 was hence named here as resistance-inducing protein of Ehrlichia (RipE). Using this Δ ripE mutant, we studied how RipE protein contributes to Ehrlichia pathogenesis. Ehrlichia species have an intracellular developmental cycle and a brief extracellular stage to initiate a new cycle of infection. Majority of RipE proteins were expressed on the surface of the smaller infectious dense-core stage of bacteria. Extracellular Δ ripE E. japonica contained significantly less adenosine triphosphate (ATP) and lost infectivity more rapidly in culture compared with wild-type (WT) E. japonica . Genetic complementation in the Δ ripE mutant or overexpression of ripE in WT E. japonica significantly increased bacterial ATP levels, and RipE-overexpressing E. japonica was more virulent in mice than WT E. japonica . RipE is conserved among Ehrlichia species. Immunization of mice with recombinant RipE induced an in vitro infection-neutralizing antibody, significantly prolonged survival time after a lethal dose of E. japonica challenge, and cross-protected mice from infection by Ehrlichia chaffeensis , the agent of human monocytic ehrlichiosis. Our findings shed light on the extracellular stage of Ehrlichia , highlighting the importance of RipE and ATP levels in Ehrlichia for extracellular resistance and the next cycle of infection. Thus, RipE is a critical Ehrlichia protein for infection as such can be a potential vaccine target for ehrlichiosis., Competing Interests: The authors declare that the research was conducted in the absence of any commercial or financial relationships that could be construed as a potential conflict of interest. The author(s) declared that they were an editorial board member of Frontiers, at the time of submission. This had no impact on the peer review process and the final decision., (Copyright © 2024 Chien, Lin, Duan, Denton, Kawahara and Rikihisa.)

Generative pretrained models have achieved remarkable success in various domains such as language and computer vision. Specifically, the combination of large-scale diverse datasets and pretrained transformers has emerged as a promising approach for developing foundation models. Drawing parallels between language and cellular biology (in which texts comprise words; similarly, cells are defined by genes), our study probes the applicability of foundation models to advance cellular biology and genetic research. Using burgeoning single-cell sequencing data, we have constructed a foundation model for single-cell biology, scGPT, based on a generative pretrained transformer across a repository of over 33 million cells. Our findings illustrate that scGPT effectively distills critical biological insights concerning genes and cells. Through further adaptation of transfer learning, scGPT can be optimized to achieve superior performance across diverse downstream applications. This includes tasks such as cell type annotation, multi-batch integration, multi-omic integration, perturbation response prediction and gene network inference., (© 2024. The Author(s), under exclusive licence to Springer Nature America, Inc.)

Objectives: Systemic lupus erythematosus (SLE) is a heterogeneous autoimmune disease. Its diagnosis poses significant challenges especially at early stages and in atypical cases. The aim of this study was to develop a machine learning model based on common laboratory tests that can aid SLE diagnosis., Methods: A standard protocol was developed to collect data of SLE and control immune diseases. A 10-fold cross-validation was performed in the modeling dataset (n=862), and an external dataset (n=198) was used for model validation. Machine learning algorithms were applied to construct a diagnostic model. Performance was evaluated based on area under the curve (AUC) values, F1-score, negative predictive value, positive predictive value, accuracy, sensitivity, and specificity., Results: The optimal model was based on a random forest algorithm with 10 clinical features. Thrombin time, prothrombin activity, and uric acid contributed most to the diagnostic model. The SLE diagnostic model showed sufficient predictive accuracy, with AUC values of 0.8286 in the validation dataset., Conclusions: Our diagnostic model based on 10 common laboratory tests identified the patients with SLE with high accuracy. An online version of the model can potentially be applied in clinical settings for the differential diagnosis of SLE.

Background: Primary Sjögren's syndrome (pSS) is a common chronic systemic autoimmune disorder which primarily affects the exocrine glands. Patients may have extraglandular disease involving multiple organs, including the kidneys. This study aimed at investigating the clinical data and laboratory markers which were associated with renal function damage or renal involvement., Method: One thousand two hundred eighty-eight adult pSS patients from the Department of Rheumatology and Clinical Immunology were enrolled in this retrospective cohort study. And there were 334 patients of them followed up for more than two years for analyzing demographic, clinical data and laboratory markers. Statistical analysis was performed by R software (Version 3.6.2)., Result: Nearly 95% of 1288 pSS patients were women, and the positive rates of anti-SSA (Sjögren's syndrome A) and anti-SSB were 63% and 27% respectively. 12% of the pSS patients presented renal involvement with eGFR < 60 mL/min/1.73 m 2 , and the mean age of hospital presentation, serum creatinine and urea were the highest (P < 0.001), and ANA (antinuclear antibody)-positive, anti-SSB-positive and anti-scl-70-positive were more prevalent in this group. Multivariate analyses showed that age, urea, chlorine and anti-SSA indicate a significant association with renal dysfunction. Potassium, sodium and Jo-1 were also confirmed to be related with decreased renal function. The receiver operating characteristic (ROC) analysis including the above factors showed a good performance on the evaluation of renal injury including eGFR < 60 mL/min/1.73 m 2 and eGFR 60 -90 mL/min/1.73 m 2 in pSS, with area under curve (AUC) values of 0.957 and 0.821, and high sensitivity (71.1% and 84.4%) and specificity (95.5% and 70.5%). After a more than two years follow-up of anti-SSA positive patients, 34.14% of them developed decreased renal function, and 13.58% of them experienced a progression of renal injury with a 23.64% decrease in eGFR., Conclusion: Age, urea, chlorine, and anti-SSA were highly associated with renal injury in pSS. Early screening for autoantibodies would be meaningful for evaluation and prevention of renal injury in pSS., (© 2023. BioMed Central Ltd., part of Springer Nature.)

Ehrlichia chaffeensis , a small Gram-negative obligatory intracellular bacterium, infects human monocytes or macrophages, and causes human monocytic ehrlichiosis, one of the most prevalent, life-threatening emerging zoonoses. Reactive oxygen species are produced by the host immune cells in response to bacterial infections. The mechanisms exploited by E. chaffeensis to resist oxidative stress have not been comprehensively demonstrated. Here, we found that E. chaffeensis encodes two functional enzymes, GshA and GshB, to synthesize glutathione that confers E. chaffeensis the oxidative stress resistance, and that the expression of gshA and gshB is upregulated by CtrA, a global transcriptional regulator, upon oxidative stress. We found that in E. chaffeensis , the expression of gshA and gshB was upregulated upon oxidative stress using quantitative RT-PCR. Ehrlichia chaffeensis GshA or GshB restored the ability of Pseudomonas aeruginosa GshA or GshB mutant to cope with oxidative stress, respectively. Recombinant E. chaffeensis CtrA directly bound to the promoters of gshA and gshB , determined with electrophoretic mobility shift assay, and activated the expression of gshA and gshB determined with reporter assay. Peptide nucleic acid transfection of E. chaffeensis , which reduced the CtrA protein level, inhibited the oxidative stress-induced upregulation of gshA and gshB . Our findings provide insights into the function and regulation of the two enzymes critical for E. chaffeensis resistance to oxidative stress and may deepen our understanding of E. chaffeensis pathogenesis and adaptation in hosts., Competing Interests: The authors declare that the research was conducted in the absence of any commercial or financial relationships that could be construed as a potential conflict of interest., (Copyright © 2022 Yan, Liang, Chai, Duan, Li, Liu, Yang, Li, Jin, Bai, Wu and Cheng.)

Because of the relatively limited understanding of coronavirus disease 2019 (COVID-19) pathogenesis, immunological analysis for vaccine development is needed. Mice and macaques were immunized with an inactivated severe acute respiratory syndrome coronavirus 2 (SARS-CoV-2) vaccine prepared by two inactivators. Various immunological indexes were tested, and viral challenges were performed on day 7 or 150 after booster immunization in monkeys. This inactivated SARS-CoV-2 vaccine was produced by sequential inactivation with formaldehyde followed by propiolactone. The various antibody responses and specific T cell responses to different viral antigens elicited in immunized animals were maintained for longer than 150 days. This comprehensive immune response could effectively protect vaccinated macaques by inhibiting viral replication in macaques and substantially alleviating immunopathological damage, and no clinical manifestation of immunopathogenicity was observed in immunized individuals during viral challenge. This candidate inactivated vaccine was identified as being effective against SARS-CoV-2 challenge in rhesus macaques., Competing Interests: The authors declare no competing interests. The corresponding authors had full access to all the data generated in the present study and assume full responsibility for the final submission of this manuscript for publication., (© 2021 The Author(s).)

Background: Bloodstream infection (BSI) is an increasing public health concern worldwide, representing a serious infection with significant morbidity and mortality, especially in children and the elderly. The predominant microbial distribution and antibiotic susceptibility were investigated among BSIs in the different intensive care units (ICUs)-pediatric ICU (PICU), surgical ICU (SICU), cardiac ICU (CICU), respiratory ICU (RICU), and geriatric ICU (GICU)-in order to achieve more efficient and appropriate therapies for patients in various ICUs. Methods: In this retrospective cross-sectional study, the blood specimens were collected from five different ICUs of Peking University First Hospital and comprehensive ICU of Miyun Teaching Hospital (Miyun ICU) before antimicrobial treatment from 2017 to 2020. Microorganism cultures of the blood samples were conducted, and positive cultures were tested for type of pathogens and drug susceptibility. Results: The prevalence of BSIs was the highest in the Miyun ICU (10.85%), followed by the RICU (9.48%) and the PICU (8.36%). The total prevalence of Gram-positive bacterial strains (especially Staphylococcus spp. and Enterococcus spp. ) in the PICU (44.55%), SICU (57.58%), CICU (55.00%), GICU (49.06%), and Miyun ICU (57.58%) was higher than that of Gram-negative bacteria. The major bacterial strain was Acinetobacter baumannii in the PICU (21.82%); Klebsiella pneumoniae in the SICU (12.88%), CICU (30.00%), and RICU (30.39%); Escherichia coli in the GICU (20.75%); and Staphylococcus epidermidis (18.18%) in the Miyun ICU. Staphylococcus hominis of BSIs remained highly susceptible (>70%) to gentamicin, linezolid, daptomycin, teicoplanin, vancomycin, tigecycline, and rifampicin in all the ICUs. Its antibiotic sensitivity to levofloxacin was moderate in the PICU and CICU, but mild (<30%) in the SICU, RICU, and GICU. K . pneumoniae was highly susceptible to doxycycline, minocycline, and tigecycline in all the ICUs except the RICU, and its antibiotic sensitivity to imipenem, meropenem, amikacin, ciprofloxacin, and levofloxacin was high/moderate in the PICU, CICU, GICU, and Miyun ICU, but mild in the SICU and RICU. Conclusion: The current study demonstrated the distribution of prevalent microorganisms, and their antimicrobial susceptibility exhibited a high divergence among BSIs in different ICUs from a tertiary hospital and an outer suburban hospital in Beijing. Therefore, different antibiotic therapies for various wards and distinct age groups (especially between pediatric and elderly patients) should be considered to control the emergence and spread of highly antibiotic-resistant infections., Competing Interests: The authors declare that the research was conducted in the absence of any commercial or financial relationships that could be construed as a potential conflict of interest., (Copyright © 2021 Duan, Sun, Huang, Li and Cheng.)

To explore an efficient strategy for further development of anticancer fluoroquinolone candidates derived from ciprofloxacin, a heterocyclic ring as the bioisosteric replacement of C3 carboxyl group led to a key intermediate, oxadiazole thiol (5), which was further modified to the bis-oxadiazole methylsulfides (7a-7h) and the corresponding dimethylpiperazinium iodides (8a-8h), respectively. Structures were characterized by elemental analysis and spectra data, and their anticancer activities in vitro against CHO, HL60 and L1210 cancer cells were also evaluated by MTT assay. The preliminary results show that piperazinium compounds (8) possess more potent activity than that of corresponding free bases (7).

An efficient modified route based on the targeting mechanism of antibacterial fluoroquinolones for the shift from the antibacterial activity to the antitumor one was further developed. Using a fused heterocyclic ring, s-triazolothiadiazine as a carboxyl bioisostere of ciprofloxacin, the title compounds, 1-cyclopropyl-6-fluoro-7-piperazin-1-yl-3-(6-substituted-phenyl-7H-[1, 2, 4]triazolo[3, 4-b][1, 3, 4]thiadiazin-3-yl)-quinolin-4(1H)-ones (5a-5e) and their corresponding N-acetyl products (6a-6e), were designed and synthesized, separately. Meaningfully, a ring-contraction of fused six-membered thiadiazine occurred by a sulfur extrusion reaction gave new tri-acetylated fused heterocycles related to pyrazolo[5, 1-c][1, 2, 4] triazoles (7a-7e). The in vitro antitumor activity against L1210, CHO and HL60 cell lines was also evaluated for the synthesized fifteen heterocycles compared to parent ciprofloxacin by methylthiazole trazolium (MTT) assay. Interestingly, the results displayed that fifteen fused heterocyclic compounds showed more significant growth inhibitory activity (IC50 < 25.0 micromo x L(-1)) than that of parent ciprofloxacin (IC50 > 150.0 micromol x L(-1)), and the active order decreased from 7a-7e to 5a-5e to 6a-6e, respective.