Sugemalimab combined with chemotherapy for the treatment of advanced esophageal squamous cell carcinoma: a cost-effectiveness analysis.

Background: This study aims to systematically analyze the cost-effectiveness of the combination therapy comprising sugemalimab and chemotherapy in the management of advanced ESCC from the Chinese healthcare system perspective. Methods: An advanced ESCC patient simulation partitioned survival approach model was developed to mimic the disease progression of patients undergoing treatment with sugemalimab in combination with chemotherapy versus chemotherapy alone. To ensure accuracy and precision, clinical data, treatment costs, and utility values were collected from comprehensive clinical trials and reliable economic databases. The cost-effectiveness analysis was conducted by assessing the incremental cost-effectiveness ratio in relation to the established willingness-to-pay threshold. One-way and probabilistic sensitivity analyses were performed to assess the robustness of the model. Results: The cumulative expenditure for the group of patients administered with sugemalimab amounted to US$ 41734.87, whereas the placebo group was associated with a total cost of US$ 22926.25. By evaluating the ICER, which quantifies the additional cost incurred per QALY gained, a value of US$ 61066.96 per QALY was determined. It is imperative to note that this ICER value surpasses the predetermined threshold for WTP in China, set at US$ 39,855.79 per QALY. Sensitivity analyses demonstrated that the results were sensitive to the cost of sugemalimab, progression-free survival, and utility values. These fluctuations did not result in a reversal of the study findings. Conclusion: The combination of sugemalimab with chemotherapy for the treatment of ESCC in China is currently not considered a cost-effective therapeutic approach. However, it is suggested that additional reductions in price may facilitate the potential for achieving cost-effectiveness.

Pembrolizumab combined with chemotherapy versus placebo combined with chemotherapy for HER2-negative advanced gastric cancer in China: a cost-effectiveness analysis.

OBJECTIVE: This study aims to conduct a cost-effectiveness analysis of pembrolizumab in combination with chemotherapy for HER2-negative advanced gastric cancer in China. METHODS: A partitioned survival approach model was constructed to simulate the progression of HER2-negative advanced gastric cancer and evaluate the outcomes of different treatment strategies. We calculated incremental cost-effectiveness ratios (ICER) to assess the cost associated with each quality-adjusted life-year (QALY) gained. One-way sensitivity analysis and probabilistic sensitivity analysis were performed to assess robustness and reliability. RESULTS: The analysis conducted in the base case demonstrated that the ICER associated with pembrolizumab was $177405.83/QALY gained in all population. In the subgroup analysis, it was found that individuals with a PD-L1 CPS ≥ 1 and those with a PD-L1 CPS ≥ 10 had ICERs of $152397.06/QALY and $109534.13/QALY, respectively. All ICER values for both the all population groups and the subgroups exceeded the WTP threshold in China. Our analysis shows the robustness of these results, as they remained consistent when input parameters were varied within a ± 25% range. CONCLUSION: The findings of this cost-effectiveness analysis suggest that pembrolizumab in combination with chemotherapy is not a cost-effective treatment option for HER2-negative advanced gastric cancer in China.

Fragment-Based Anti-inflammatory Agent Design and Target Identification: Discovery of AF-45 as an IRAK4 Inhibitor to Treat Ulcerative Colitis and Acute Lung Injury.

UC and ALI are inflammatory diseases with limited treatment in the clinic. Herein, fragment-based anti-inflammatory agent designs were carried out deriving from cyclohexylamine/cyclobutylamine and several fragments from anti-inflammatory agents in our lab. AF-45 (IC50 = 0.53/0.60 µM on IL-6/TNF-α in THP-1 macrophages) was identified as the optimal molecule using ELISA and MTT assays from the 33 synthesized compounds. Through mechanistic studies and a systematic target search process, AF-45 was found to block the NF-κB/MAPK pathway and target IRAK4, a promising target for inflammation and autoimmune diseases. The selectivity of AF-45 targeting IRAK4 was validated by comparing its effects on other kinase/nonkinase proteins. In vivo, AF-45 exhibited a good therapeutic effect on UC and ALI, and favorable PK proprieties. Since there are currently no clinical or preclinical trials for IRAK4 inhibitors to treat UC and ALI, AF-45 provides a new lead compound or candidate targeting IRAK4 for the treatment of these diseases.

Trends and age-period-cohort effect on incidence of hepatitis B from 2008 to 2022 in Guangzhou, China.

Hepatitis B virus (HBV) infection is highly prevalent in Guangzhou, China. This study aimed to examine the long-term trend of HB incidence from 2008 to 2022 and the independent impacts of age, period, and cohort on the trends. HBV data were collected from the China Information System for Disease Control and Prevention. Joinpoint regression was utilized to examine temporal trends, and an age-period-cohort model was employed to estimate the effects of age, period, and cohort. A total of 327,585 HBV cases were included in this study. The incidence of chronic and acute HB showed a decreasing trend in Guangzhou over the past 15 years, with an average annual percent change of - 4.31% and - 16.87%, respectively. Age, period, and cohort all exerted significant effects. The incidence of HB was higher in males than in females and non-central areas compared to central areas. Age groups of 0-4 years and 15-24 years were identified as high-risk groups. The period relative risks for chronic HB incidence decreased initially and then stabilized. Cohorts born later had lower risks. Chronic HB incidences remain high in Guangzhou, especially among males, younger individuals, and residents of non-central areas. More efforts are still needed to achieve hepatitis elimination targets.

Characterization of Acinetobacter baumannii at a tertiary hospital in Guangzhou: a genomic and clinical study.

Acinetobacter baumannii (AB) infections have become a global public health concern due to the continued increase in the incidence of infection and the rate of resistance to carbapenems. This study aimed to investigate the genomic features of AB strains recovered from a tertiary hospital and assess the clinical implications of the findings. A total of 217 AB strains were collected between 2016 and 2018 at a tertiary hospital in Guangzhou, with 183 (84.33%) being carbapenem-resistant AB (CRAB), with the main mechanism being the carriage of the blaOXA-23 gene. The overall mortality rate of patients caused by such strains was 15.21% (n = 33). Artificial lung ventilation and the use of meropenem were mortality risk factors in AB-infected patients, while KL2 AB infection was negatively associated. Core genome multilocus sequence typing and clustering analysis were performed on the integrated AB genome collection from the NCBI database and this study to illustrate the population structure among China. The results revealed diverse core genome profiles (n = 17) among AB strains from China, and strains from this single hospital exhibited most of the core genome profiles (n = 13), suggesting genetic variability within the hospital and transmission across the country. These findings show that the high transmission potential of the CRAB strains and meropenem usage that confers a selective advantage of CRAB clinically are two major factors that pose significant challenges to the effective clinical management of AB infections. Understanding the genetic features and transmission patterns of clinical AB strains is crucial for the effective control of infections caused by this pathogen.

A comparative study of large language model-based zero-shot inference and task-specific supervised classification of breast cancer pathology reports.

OBJECTIVE: Although supervised machine learning is popular for information extraction from clinical notes, creating large annotated datasets requires extensive domain expertise and is time-consuming. Meanwhile, large language models (LLMs) have demonstrated promising transfer learning capability. In this study, we explored whether recent LLMs could reduce the need for large-scale data annotations. MATERIALS AND METHODS: We curated a dataset of 769 breast cancer pathology reports, manually labeled with 12 categories, to compare zero-shot classification capability of the following LLMs: GPT-4, GPT-3.5, Starling, and ClinicalCamel, with task-specific supervised classification performance of 3 models: random forests, long short-term memory networks with attention (LSTM-Att), and the UCSF-BERT model. RESULTS: Across all 12 tasks, the GPT-4 model performed either significantly better than or as well as the best supervised model, LSTM-Att (average macro F1-score of 0.86 vs 0.75), with advantage on tasks with high label imbalance. Other LLMs demonstrated poor performance. Frequent GPT-4 error categories included incorrect inferences from multiple samples and from history, and complex task design, and several LSTM-Att errors were related to poor generalization to the test set. DISCUSSION: On tasks where large annotated datasets cannot be easily collected, LLMs can reduce the burden of data labeling. However, if the use of LLMs is prohibitive, the use of simpler models with large annotated datasets can provide comparable results. CONCLUSIONS: GPT-4 demonstrated the potential to speed up the execution of clinical NLP studies by reducing the need for large annotated datasets. This may increase the utilization of NLP-based variables and outcomes in clinical studies.

Design, synthesis and bioactivity evaluation of 4-hydroxycoumarin derivatives as potential anti-inflammatory agents against acute lung injury and colitis.

Acute lung injury (ALI) and inflammatory bowel disease (IBD) are common inflammatory illnesses that seriously affect people's health. Herein, a series of 4-hydroxylcoumarin (4-HC) derivatives were designed and synthesized. The inhibitory effects of these compounds on LPS-induced interleukin-6 (IL-6) release from J774A.1 cells were then screened via ELISA assay, compound B8 showed 3 times more active than the lead compound 4-HC. The most active compound B8 had the IC50 values of 4.57 µM and 6.51 µM for IL-6 release on mouse cells J774A.1 and human cells THP-1, respectively. Furthermore, we also found that B8 could act on the MAPK pathway. Based on the target prediction results of computer virtual docking, kinase inhibitory assay was carried out, and it revealed that targeting IRAK1 was a key mechanism for B8 to exert anti-inflammatory activity. Moreover, B8 exerted a good therapeutic effect on the dextran sulfate sodium (DSS)-induced colitis model and liposaccharide (LPS)-induced ALI mouse models. The acute toxicity experiments indicated that high-dose B8 caused no adverse reactions in mice, confirming its safety in vivo. Additionally, the preliminary pharmacokinetic (PK) parameters of B8 in SD rats were also examined, revealing a bioavailability (F) of 28.72 %. In conclusion, B8 is a potential candidate of drug for the treatment of ALI and colitis.

Metal Hydrogen-Bonded Organic Frameworks as Open Lewis Acid Catalysts for Two Types of CO2 Transformations.

Efficient and multiple CO2 utilization into high-value-added chemicals holds significant importance in carbon neutrality and industry production. However, most catalysis systems generally exhibit only one type of CO2 transformation with the efficiency to be improved. The restricted abundance of active catalytic sites or an inefficient utilization rate of these sites results in the constraint. Consequently, we designed and constructed two metal hydrogen-bonded organic frameworks (M-HOFs) {[M3(L3-)2(H2O)10]·2H2O}n (M = Co (1), Ni (2); L = 1-(4-carboxyphenyl)-1H-pyrazole-3,5-dicarboxylic acid) in this research. 1 and 2 are well-characterized, and both show excellent stability. The networks connected by multiple hydrogen bonds enhance the structural flexibility and create accessible Lewis acidic sites, promoting interactions between the substrates and catalytic centers. This enhancement facilitates efficient catalysis for two types of CO2 transformations, encompassing both cycloaddition reactions with epoxides and aziridines to afford cyclic carbonates and oxazolidinones. The catalytic activities (TON/TOF) are superior compared with those of most other catalysts. These heterogeneous catalysts still exhibited high performance after being reused several times. Mechanistic studies indicated intense interactions between the metal sites and substrates, demonstrating the reason for efficient catalysis. This marks the first instance on M-HOFs efficiently catalyzing two types of CO2 conversions, finding important significance for catalyst design and CO2 utilization.

Study on the Mechanism of Yadanzi Oil in Treating Lung Cancer Based on Network Pharmacology and Molecular Docking Technology.

Brucea javanica oil emulsion (BJOE) is a compound Chinese medicine used for treating various cancers, such as lung cancer. However, the exact mechanism of its antilung cancer active ingredient remains unclear. This study aims to explore and validate the effective active ingredients and mechanism of action of BJOE in the treatment of lung cancer through network pharmacology, molecular docking technology, and cell experiments. The results showed that there were 13 active ingredients, 136 target genes, and 42 disease target-coexpressed genes in BJOE. The molecular docking results indicated that the main active components of the oil emulsion, YD1 (ß-sitosterol), YD2 (luteolin), and YD3 (bruceitol), could stably bind to TP53 and MAPK1. Furthermore, the commercially available ß-sitosterol luteolin was used as a representative compound to conduct cell experiments to verify its anticancer activity and mechanism. It was found that luteolin can inhibit the proliferation better than ß-sitosterol and the activity of lung cancer cells by regulating the expression of related proteins through the P53/MAPK1 signaling pathway. This study combines network pharmacology prediction with experiments to demonstrate the "multicomponent, multitarget, multipathway" approach of B. javanica oil emulsion in treating lung cancer.

Epidemiological and genetic characterization of multidrug-resistant non-O1 and non-O139 Vibrio cholerae from food in southern China.

This study reports a comprehensive epidemiological and genetic analysis of V. cholerae strains, specifically non-O1/non-O139 serogroups, isolated from animal-derived food samples in Guangdong province from 2015 to 2019. A total of 21 V. cholerae strains were obtained, which exhibited high resistance rates for nalidixic acid (57.14 %, 12/21), ampicillin (33.33 %, 7/21), and ciprofloxacin (19.05 %, 4/21). The quinolone resistance-related gene, qnrVC, was prevalent in 80.95 % (17/21) of the isolates. Additionally, chromosomally mediated quinolone-resistance mutations, including mutations in GyrA at position 83 (S83I) and ParC at position 85 (S85L), were detected in 47.62 % of the isolates. The combination of target mutation and qnrVC genes was shown to mediate resistance or intermediate resistance to ciprofloxacin in V. cholerae. Furthermore, an IncC-type conjugative plasmid carrying thirteen antibiotic resistance genes, including genes conferring resistance to two clinically important antibiotics, cephalosporins and fluoroquinolones, was identified in the shrimp-derived strain Vc516. While none of our food isolates harbored the toxigenic CTX- and TCP-encoding genes, they did possess genes encoding toxins such as HlyA and Autoinducer-2. Notably, some V. cholerae strains from this study exhibited a close genetic relationship with clinical strains, suggesting their potential to cause human infections. Taken together, this study provides a comprehensive view of the epidemiological features and genetic basis of antimicrobial resistance and virulence potential of V. cholerae strains isolated from food in southern China, thereby advancing our understanding of this important pathogen.

Effects of tumor marker regression load score on long-term prognosis of gastric cancer patients undergoing radical surgery after neoadjuvant chemotherapy.

BACKGROUND: The effects of the dynamics of serum tumor markers (CA72-4, CEA, CA19-9, CA125 and AFP) before and after neoadjuvant chemotherapy (NACT) on the prognosis of gastric cancer(GC) patients remain unclear. METHODS: The training set contained 334 GC patients from Fujian Medical University Union Hospital (FJMUUH) and 113 GC patients in Qinghai University Affiliated Hospital (QhUAH) were used as an external validation set. Tumor marker regression load (ΔTMRL) indicator, including ΔCA72-4, ΔCEA, ΔCA19-9, ΔCA125, and ΔAFP, is defined as [(postNACT marker- preNACT marker)/preNACT marker]. Tumor marker regression load score (TMRLS) consists of ΔCA72-4, ΔCEA and ΔCA125. The predictive performance of the nomogram-TMRLS was evaluated using the area under the receiver operating characteristic(ROC) curve(AUC), decision curve analysis(DCA), and C-index. RESULTS: Patients from FJMUUH were divided into two groups, TMRLS-low and TMRLS-high, determined by R package maxstat. Survival analysis revealed a higher 3-year overall survival(OS) in the TMRLS-low than in the TMRLS-high group. The TMRLS-high group who received postoperative adjuvant chemotherapy(AC) showed a significantly higher 3-year OS rate than those who did not. Multivariate COX regression analysis indicated that TMRLS was an independent prognostic factor for OS. A nomogram for predicting OS based on TMRLS showed a significantly higher C-index and AUC than the ypTNM stage. The above results were also found in the QhUAH external validation cohort. CONCLUSION: TMRLS is a novel independent prognostic factor for GC who underwent NACT and a radical gastrectomy. Furthermore, the TMRLS-high group, who received postoperative AC, may achieve better survival outcomes. Notably, the predictive performance of the nomogram-TMRLS significantly outperformed that of the ypTNM stage.

Tislelizumab versus sorafenib as first-line treatment for advanced hepatocellular carcinoma in China: a cost-effectiveness analysis.

Objective: The goal of this study is to compare the cost-effectiveness of tislelizumab and sorafenib as first-line treatment for advanced hepatocellular carcinoma in China. Methods: A comprehensive cost-effectiveness analysis was undertaken within the framework of a partitioned survival model to accurately gage the incremental cost-effectiveness ratio (ICER) of tislelizumab compared to sorafenib. The model incorporated relevant clinical data and all survival rates were from RATIONALE-301 trials. The stability of the partitioned survival model was assessed by performing one-way and two-way sensitivity analyses. Results: The total cost incurred for the tislelizumab treatment was $16181.24, whereas the sorafenib was $14306.87. The tislelizumab regimen resulted in a significant increase of 0.18 quality-adjusted life years (QALYs) and an extra cost of $1874.37 as compared to chemotherapy. The ICER was $10413.17 per QALY, which was found to be below the willingness-to-pay (WTP) threshold of $37304.34/QALY. The results of the sensitivity analysis found that no fluctuations in any of the factors affected our results, even when these parameters fluctuated. Conclusion: Tislelizumab appears to be a cost-effective first-line treatment for advanced hepatocellular carcinoma when compared to sorafenib in China. These findings can inform decision-making processes regarding the selection of the most cost-effective treatment option for advanced hepatocellular carcinoma.

Quantum chemistry calculation-aided discovery of potent small-molecule mimics of glutathione peroxidases for the treatment of cisplatin-induced hearing loss.

Hearing loss (HL) is a health burden that seriously affects the quality of life of cancer patients receiving platinum-based chemotherapy, and few FDA-approved treatment specifically targets this condition. The main mechanisms that contribute to cisplatin-induced hearing loss are oxidative stress and subsequent cell death, including ferroptosis revealed by us as a new mechanism recently. In this study, we employed the frontier molecular orbital (FMO) theory approach as a convenient prediction method for the glutathione peroxidase (GPx)-like activity of isoselenazolones and discovered new isoselenazolones with great GPx-like activity. Notably, compound 19 exhibited significant protective effects against cisplatin-induced hair cell (HC) damage in vitro and in vivo and effectively reverses cisplatin-induced hearing loss through oral administration. Further investigations revealed that this compound effectively alleviated hair cell oxidative stress, apoptosis and ferroptosis. This research highlights the potential of GPx mimics as a therapeutic strategy against cisplatin-induced hearing loss. The application of quantum chemistry (QC) calculations in the study of GPx mimics sheds light on the development of new, innovative treatments for hearing loss.

Redundancy effect of initial enactment encoding and subsequent testing on memory enhancement: Insights from an electrophysiological study.

Testing is more beneficial for memory retention than restudying the same content. However, the effect of the initial encoding method on the testing effect remains unclear. In this study, a classical testing effect paradigm was employed, along with event-related potentials (ERP), to investigate the electrophysiological processes underlying the effect of enactment encoding on the testing effect. Participants were randomly assigned to the Self-Performed Test (SPT) or Verbalized Test (VT) groups. Both groups underwent three stages: an initial encoding phase, an initial test phase (comprising a source memory task and a restudy task), and a final test phase. During the initial encoding phase, the SPT group encoded action phrases through enactment, while the VT group encoded information through reading. During the initial test phase, the SPT group exhibited superior recognition performance in item memory compared with the VT group. Both groups exhibited significant parietal old/new effects in the source memory task, with only the SPT group displaying parietal positivity during the restudy task. During the final test phase, the behavioral testing effect was exclusively observed in the VT group. Furthermore, the VT group displayed a more pronounced parietal positivity in the test condition compared to the restudy condition, while the parietal positivity between the two conditions was comparable in the SPT group. In summary, the absence of a final behavioral testing effect in the SPT group may be attributed to both enactment and testing primarily enhancing memory performance through recollection-based retrieval, as indicated by the parietal positivity. Consequently, the initial enactment encoding method leaves limited scope for further improvements through subsequent testing. These findings suggest that initial enactment encoding, and subsequent testing may be redundant in improving episodic memory performance.

Robotic gastrectomy was reliable option for overweight patients with gastric cancer: a propensity score matching study.

BACKGROUND: The role of minimally invasive surgery using robotics versus laparoscopy in resectable gastric cancer patients with a high body mass index (BMI) remains controversial. METHODS: A total of 482 gastric adenocarcinoma patients with BMI ≥ 25 kg/m2 who underwent minimally invasive radical gastrectomy between August 2016 and December 2019 were retrospectively analyzed, including 109 cases in the robotic gastrectomy (RG) group and 321 cases in the laparoscopic gastrectomy (LG) group. Propensity score matching (PSM) with a 1:1 ratio was performed, and the perioperative outcomes, lymph node dissection, and 3-year overall survival (OS) and disease-free survival (DFS) rates were compared. RESULTS: After PSM, 109 patients were included in each of the RG and LG groups, with balanced baseline characteristics. Compared with the LG group, the RG group had similar intraoperative estimated blood loss [median (IQR) 30 (20-50) vs. 35 (30-59) mL, median difference (95%CI) - 5 (- 10 to 0)], postoperative complications [13.8% vs. 18.3%, OR (95%CI) 0.71 (0.342 to 1.473)], postoperative recovery, total harvested lymph nodes [(34.25 ± 13.43 vs. 35.44 ± 14.12, mean difference (95%CI) - 1.19 (- 4.871 to 2.485)] and textbook outcomes [(81.7% vs. 76.1%, OR (95%CI) 1.39 (0.724 to 2.684)]. Among pathological stage II-III patients receiving chemotherapy, the initiation of adjuvant chemotherapy in the RG group was similar to that in the LG group [median (IQR): 28 (25.5-32.5) vs. 32 (27-38.5) days, median difference (95%CI) - 3 (- 6 to 0)]. The 3-year OS (RG vs. LG: 80.7% vs. 81.7%, HR = 1.048, 95%CI 0.591 to 1.857) and DFS (78% vs. 76.1%, HR = 0.996, 95%CI 0.584 to 1.698) were comparable between the two groups. CONCLUSION: RG conferred comparable lymph node dissection, postoperative recovery, and oncologic outcomes in a selected cohort of patients with BMI ≥ 25 kg/m2.

[Quantitative Assessment of the Impact of Climate Change on the Growing Season of Vegetation Gross Primary Productivity in the Middle and Lower Reaches of the Yangtze River].

Quantitatively determining the direct, indirect, and comprehensive effects of climatic factors on the growing season of the vegetation GPP (GPPGS) in the middle and lower reaches of the Yangtze River at the regional and vegetation type scales can provide a scientific basis for the management and restoration of regional vegetation resources under the background of global climate change. Using MODIS GPP data, meteorological data, and vegetation type data, combined with Theil-Sen Median trend analysis and the Mann-Kendall significance test, the spatiotemporal characteristics of the GPPGS in the middle and lower reaches of the Yangtze River were investigated at different temporal and spatial scales. Path analysis was used to further reveal the direct, indirect, and comprehensive effects of climate factors on GPPGS variation in different vegetation types. The results showed that:① from 2000 to 2021, the vegetation GPPGS in the middle and lower reaches of the Yangtze River showed a fluctuating upward trend, with a rising rate (in terms of C, same below) of 2.70 g·(m2·a)-1 (P<0.01). The GPPGS of different vegetation types all showed a significant upward trend (P<0.01), with shrubs having the highest upward rate of 3.31 g·(m2·a)-1 and cultivated vegetation having the lowest upward rate of 2.54 g·(m2·a)-1. ② The proportion of the area with an upward trend in GPPGS in the middle and lower reaches of the Yangtze River was 88.11%. The proportion of the area with an upward trend in GPPGS was greater than 84% for all different vegetation types, with shrubs (49.76%) and cultivated vegetation (44.36%) having significantly higher proportions of the area with an upward trend than that in other vegetation types. ③ The path analysis results showed that precipitation and the maximum temperature had a significant positive direct effect on vegetation GPPGS (P<0.05), whereas solar radiation had a non-significant positive effect (P ≥ 0.05). The indirect effects of maximum temperature, precipitation, and solar radiation on vegetation GPPGS were all non-significantly negative (P ≥ 0.05). Under the combined effects of direct and indirect influences, precipitation and maximum temperature had a non-significant positive effect on vegetation GPPGS (P ≥ 0.05), whereas solar radiation had a non-significant negative effect on vegetation GPPGS (P ≥ 0.05). Among different vegetation types, precipitation was the main climate factor affecting the changes in GPPGS of cultivated vegetation, whereas the maximum temperature was the main climate factor affecting the changes in GPPGS of coniferous forests, broad-leaved forests, shrubs, and grasslands. ④ The changes in vegetation GPPGS in the middle and lower reaches of the Yangtze River were mainly influenced by the direct effects of maximum temperature, precipitation, and solar radiation, with the direct effect of precipitation dominating 56.72% of the changes in GPPGS. The research results can provide a reference for quantifying the carbon sequestration potential of vegetation in the middle and lower reaches of the Yangtze River and formulating ecological restoration governance policies tailored to local conditions under the background of global climate change.

Comprehensive genomic and plasmid characterization of multidrug-resistant bacterial strains by R10.4.1 nanopore sequencing.

The escalating prevalence of multidrug-resistant (MDR) bacteria pose a significant public health threat. Understanding the genomic features and deciphering the antibiotic resistance profiles of these pathogens is crucial for development of effective surveillance and treatment strategies. In this study, we employed the R10.4.1 nanopore sequencing technology, specifically through the use of the MinION platform, to analyze eight MDR bacterial strains originating from clinical, ecological and food sources. A single 72-hour sequencing run could yield approximately 12 million reads which covered a total of 34 gigabases (Gbp). The nanopore R10.4.1 data was processed using the Flye assembler, successfully assembling the genomes of eight bacterial strains and their 18 plasmids. Notably, the assemblies generated solely from R10.4.1 nanopore data closely matched those from next-generation sequencing data. Diverse antibiotic resistance patterns and specific resistance genes in the test strains were identified. Hospital strains that exhibited multidrug resistance were found to harbor various resistance genes that encode efflux pumps and extended-spectrum ß-lactamases. Environmental and food sources were found to display resistance profiles in a species-specific manner. The composition of structurally complex plasmids in the test strains could also be revealed by analysis of nanopore long reads, which also suggested evidence of horizontal transfer of plasmids between different bacterial species. These findings provide valuable insights into the genetic characteristics of MDR bacteria and demonstrating the practicality of nanopore sequencing technology for detecting of resistance elements in bacterial pathogens.

Integrating metagenomic and isolation strategies revealed high contamination of pathogenies and resistome in market shrimps.

This study employs a comprehensive approach combining metagenomic analysis and bacterial isolation to elucidate the microbial composition, antibiotic resistance genes (ARGs), and virulence factors (VFGs) present in shrimps from market and supermarket. Metagenomic analysis of shrimps revealed a dominance of Proteobacteria and Bacteroidetes with Firmicutes notably enriched in some samples. On the other hand, the dominant bacteria isolated included Citrobacter portucalensis, Escherichia coli, Salmonella enterica, Vibrio species and Klebsiella pneumonaie. Metagenomic analysis unveiled a diverse spectrum of 23 main types and 380 subtypes of ARGs in shrimp samples including many clinical significant ARGs such as blaKPC, blaNDM, mcr, tet(X4) etc. Genomic analysis of isolated bacterial strains identified 14 ARG types with 109 subtype genes, which complemented the metagenomic data. Genomic analysis also allowed us to identify a rich amount of MDR plasmids, which provided further insights into the dissemination of resistance genes in different species of bacteria in the same samples. Examination of VFGs and mobile genetic elements (MGEs) in both metagenomic and bacterial genomes revealed a complex landscape of factors contributing to bacterial virulence and genetic mobility. Potential co-occurrence patterns of ARGs and VFGs within human pathogenic bacteria underlined the intricate interplay between antibiotic resistance and virulence. In conclusion, this integrated analysis for the first time provides a comprehensive view and sheds new light on the potential hazards associated with shrimp products in the markets. The findings underscore the necessity of ongoing surveillance and intervention strategies to mitigate risks posed by antibiotic-resistant bacteria in the food supply chain using the novel comprehensive approaches.

A comparative study of zero-shot inference with large language models and supervised modeling in breast cancer pathology classification.

Although supervised machine learning is popular for information extraction from clinical notes, creating large, annotated datasets requires extensive domain expertise and is time-consuming. Meanwhile, large language models (LLMs) have demonstrated promising transfer learning capability. In this study, we explored whether recent LLMs can reduce the need for large-scale data annotations. We curated a manually labeled dataset of 769 breast cancer pathology reports, labeled with 13 categories, to compare zero-shot classification capability of the GPT-4 model and the GPT-3.5 model with supervised classification performance of three model architectures: random forests classifier, long short-term memory networks with attention (LSTM-Att), and the UCSF-BERT model. Across all 13 tasks, the GPT-4 model performed either significantly better than or as well as the best supervised model, the LSTM-Att model (average macro F1 score of 0.83 vs. 0.75). On tasks with a high imbalance between labels, the differences were more prominent. Frequent sources of GPT-4 errors included inferences from multiple samples and complex task design. On complex tasks where large annotated datasets cannot be easily collected, LLMs can reduce the burden of large-scale data labeling. However, if the use of LLMs is prohibitive, the use of simpler supervised models with large annotated datasets can provide comparable results. LLMs demonstrated the potential to speed up the execution of clinical NLP studies by reducing the need for curating large annotated datasets. This may increase the utilization of NLP-based variables and outcomes in observational clinical studies.