Machine-learning developed an iron, copper, and sulfur-metabolism associated signature predicts lung adenocarcinoma prognosis and therapy response.

BACKGROUND: Previous studies have largely neglected the role of sulfur metabolism in LUAD, and no study has combine iron, copper, and sulfur-metabolism associated genes together to create prognostic signatures. METHODS: This study encompasses 1564 LUAD patients, 1249 NSCLC patients, and over 10,000 patients with various cancer types from diverse cohorts. We employed the R package ConsensusClusterPlus to separate patients into different ICSM (Iron, Copper, and Sulfur-Metabolism) subtypes. Various machine-learning methods were utilized to develop the ICSMI. Enrichment analyses were conducted using ClusterProfiler and GSVA, while IOBR quantified immune cell infiltration. GISTIC2.0 and maftools were utilized for CNV and SNV data analysis. The Oncopredict package predicted drug information based on GDSC1. TIDE algorithm and cohorts GSE91061 and IMvigor210 evaluated patient response to immunotherapy. Single-cell data was processed using the Seurat package, AUCell package calculated cells geneset activity scores, and the Scissor algorithm identified ICSMI-associated cells. In vitro experiments was conducted to explore the role of ICSMRGs in LUAD. RESULTS: Unsupervised clustering identified two distinct ICSM subtypes of LUAD, each with unique clinical characteristics. The ICSMI, comprising 10 genes, was constructed using integrated machine-learning methods. Its prognostic power was validated in 10 independent datasets, revealing that LUAD patients with higher ICSMI levels had poorer prognoses. Furthermore, ICSMI demonstrated superior predictive abilities compared to 102 previously published signatures. A nomogram incorporating ICSMI and clinical features exhibited high predictive performance. ICSMI positively correlated with patients gene mutations, and integrated analysis of bulk and single-cell transcriptome data revealed its association with TME modulators. Cells representing the high-ICSMI phenotype exhibited more malignant features. LUAD patients with high ICSMI levels exhibited sensitivity to chemotherapy and targeted therapy but displayed resistance to immunotherapy. In a comprehensive analysis across various cancers, ICSMI retained significant prognostic value and emerged as a risk factor for the majority of cancer patients. CONCLUSIONS: ICSMI provides critical prognostic insights for LUAD patients, offering valuable insights into the tumor microenvironment and predicting treatment responsiveness.

Identification of a novel ADCC-related gene signature for predicting the prognosis and therapy response in lung adenocarcinoma.

BACKGROUND: Previous studies have largely neglected the role of ADCC in LUAD, and no study has systematically compiled ADCC-associated genes to create prognostic signatures. METHODS: In this study, 1564 LUAD patients, 2057 NSCLC patients, and more than 5000 patients with various cancer types from diverse cohorts were included. R package ConsensusClusterPlus was utilized to classify patients into different subtypes. A number of machine-learning algorithms were used to construct the ADCCRS. GSVA and ClusterProfiler were used for enrichment analyses, and IOBR was used to quantify immune cell infiltration level. GISTIC2.0 and maftools were used to analyze the CNV and SNV data. The Oncopredict package was used to predict drug information based on the GDSC1. Three immunotherapy cohorts were used to evaluate patient response to immunotherapy. The Seurat package was used to process single-cell data, the AUCell package was used to calculate cells' geneset activity scores, and the Scissor algorithm was used to identify ADCCRS-associated cells. RESULTS: Through unsupervised clustering, two distinct subtypes of LUAD were identified, each exhibiting distinct clinical characteristics. The ADCCRS, consisted of 16 genes, was constructed by integrated machine-learning methods. The prognostic power of ADCCRS was validated in 28 independent datasets. Further, ADCCRS shows better predictive abilities than 102 previously published signatures in predicting LUAD patients' survival. A nomogram incorporating ADCCRS and clinical features was constructed, demonstrating high predictive performance. ADCCRS positively correlates with patients' gene mutation, and integrated analysis of bulk and single-cell transcriptome data revealed the association of ADCCRS with TME modulators. Cells representing high-ADCCRS phenotype exhibited more malignant features. LUAD patients with high ADCCRS levels exhibited sensitivity to chemotherapy and targeted therapy, while displaying resistance to immunotherapy. In pan-cancer analysis, ADCCRS still exhibited significant prognostic value and was found to be a risk factor for most cancer patients. CONCLUSIONS: ADCCRS offers a critical prognostic insight for patients with LUAD, shedding light on the tumor microenvironment and forecasting treatment responsiveness.

Performance of public drinking water purifiers in control of trihalomethanes, antibiotics and antibiotic resistance genes.

Point-of-use water purifiers are widely applied as a terminal treatment device to produce drinking water with high quality. However, concerns are raised regarding low efficiency in eliminating emerging organic pollutants. To enhance our understanding of the reliability and potential risks of water purifiers, the removal of trihalomethanes, antibiotics, and antibiotic resistance genes (ARGs) in four public water purifiers was investigated. In the four public water purifiers in October and November, the removal efficiencies of trichloromethane (TCM) and bromodichloromethane (BDCM) were 15%-69% (averagely 37%) and 6%-44% (averagely 23%). The levels of TCM and BDCM were lowered by all water purifiers in October and November, but accelerated in effluent compared to the influent in one public water purifier in December. The removal efficiencies of twelve antibiotics greatly varied with species and time. Out of twelve sampling cases, the removal efficiencies of total antibiotics were 25%-75% in ten cases. In the other two cases, very low removal efficiency (6%) or higher levels of antibiotics present in effluent compared to the influent were observed. Two public water purifiers effectively remove ARGs from water, with log removal rates of 0.45 log-3.89 log. However, in the other two public water purifiers, the ARG abundance accidently increased in the effluents. Overall, public water purifiers were more effective in removing antibiotics and ARGs compared to household water purifiers, but less or equally effective in removing trihalomethanes. Both public and household water purifiers could be contaminated and release the accumulated micro-pollutants or biofilm-related pollutants into effluent. The production frequency and standing time of water within water purifiers can impact the internal contamination and purification efficacy.

Effective Degradation of Free Gossypol in Defatted Cottonseed Meal by Bacterial Laccases: Performance and Toxicity Analysis.

Cottonseed meal (CSM) is the major by-product of the cottonseed oil extraction process with high protein content. However, the presence of free gossypol (FG) in CSM severely restricts its utilization in the food and animal feed industries. The development of a biological strategy for the effective removal of FG in CSM has become an urgent need. In this study, three bacterial laccases including CotA from Bacillus licheniformis, CueO from Escherichia coli, and LcLac from Loigolactobacillus coryniformis were heterologously expressed and investigated for their FG degradation ability. The results showed that CotA laccase displayed the highest FG-degrading capacity among the three laccases, achieving 100% FG degradation at 37 °C and pH 7.0 in 1 h without the addition of a redox mediator. Moreover, in vitro and in vivo studies confirmed that the hepatotoxicity of FG was effectively eliminated after oxidative degradation by CotA laccase. Furthermore, the addition of CotA laccase could achieve 87% to 98% FG degradation in defatted CSM within 2 h. In conclusion, CotA laccase can be developed as an effective biocatalyst for the detoxification of FG in CSM.

PLK1 phosphorylation of ZW10 guides accurate chromosome segregation in mitosis.

Stable transmission of genetic information during cell division requires faithful chromosome segregation. Mounting evidence has demonstrated that PLK1 dynamics at kinetochores control correct kinetochore-microtubule attachments and subsequent silencing of the spindle checkpoint. However, the mechanisms underlying PLK1-mediated silencing of the spindle checkpoint remain elusive. Here, we identified a regulatory mechanism by which PLK1-elicited ZW10 phosphorylation regulates spindle checkpoint silencing in mitosis. ZW10 is a cognate substrate of PLK1, and the phosphorylation of ZW10 at Ser12 enables dynamic ZW10-Zwint1 interactions. Inhibition of ZW10 phosphorylation resulted in misaligned chromosomes, while persistent expression of phospho-mimicking ZW10 mutant caused premature anaphase, in which sister chromatids entangled as cells entered anaphase. These findings reveal the previously uncharacterized PLK1-ZW10 interaction through which dynamic phosphorylation of ZW10 fine-tunes accurate chromosome segregation in mitosis.

Semi-Infinitely Constrained Markov Decision Processes and Provably Efficient Reinforcement Learning.

We propose a novel generalization of constrained Markov decision processes (CMDPs) that we call the semi-infinitely constrained Markov decision process (SICMDP). Particularly, we consider a continuum of constraints instead of a finite number of constraints as in the case of ordinary CMDPs. We also devise two reinforcement learning algorithms for SICMDPs that we refer to as SI-CMBRL and SI-CPO. SI-CMBRL is a model-based reinforcement learning algorithm. Given an estimate of the transition model, we first transform the reinforcement learning problem into a linear semi-infinitely programming (LSIP) problem and then use the dual exchange method in the LSIP literature to solve it. SI-CPO is a policy optimization algorithm. Borrowing ideas from the cooperative stochastic approximation approach, we make alternative updates to the policy parameters to maximize the reward or minimize the cost. To the best of our knowledge, we are the first to apply tools from semi-infinitely programming (SIP) to solve constrained reinforcement learning problems. We present theoretical analysis for SI-CMBRL and SI-CPO, identifying their iteration complexity and sample complexity. We also conduct extensive numerical experiments to illustrate the SICMDP model and demonstrate that our proposed algorithms are able to solve complex control tasks leveraging modern deep reinforcement learning techniques.

Enzymatic characterization and application of soybean hull peroxidase as an efficient and renewable biocatalyst for degradation of zearalenone.

Zearalenone (ZEN) is a notorious mycotoxin commonly found in Fusarium-contaminated crops, which causes great loss in livestock farming and serious health problems to humans. In the present work, we found that crude peroxidase extraction from soybean hulls could use H2O2 as a co-substate to oxidize ZEN. Molecular docking and dynamic simulation also supported that ZEN could bind to the active site of soybean hull peroxidase (SHP). Subsequently, SHP extracted from soybean hulls was purified using a combined purification protocol involving ammonium sulfate precipitation, ion exchange chromatography and size exclusion chromatography. The purified SHP showed wide pH resistance and high thermal stability. This peroxidase could degrade 95 % of ZEN in buffer with stepwise addition of 100 µM H2O2 in 1 h. The two main ZEN degradation products were identified as 13-OH-ZEN and 13-OH-ZEN-quinone. Moreover, SHP-catalyzed ZEN degradation products displayed much less cytotoxicity to human liver cells than ZEN. The application of SHP in various food matrices obtained 54 % to 85 % ZEN degradation. The findings in this study will promote the utilization of SHP as a cheap and renewable biocatalyst for degrading ZEN in food.

In-depth single-cell and bulk-RNA sequencing developed a NETosis-related gene signature affects non-small-cell lung cancer prognosis and tumor microenvironment: results from over 3,000 patients.

Background: Cell death caused by neutrophil extracellular traps (NETs) is known as NETosis. Despite the increasing importance of NETosis in cancer diagnosis and treatment, its role in Non-Small-Cell Lung Cancer (NSCLC) remains unclear. Methods: A total of 3298 NSCLC patients from different cohorts were included. The AUCell method was used to compute cells' NETosis scores from single-cell RNA-sequencing data. DEGs in sc-RNA dataset were obtained by the Seurat's "FindAllMarkers" function, and DEGs in bulk-RNA dataset were acquired by the DESeq2 package. ConsensusClusterPlus package was used to group patients into different NETosis subtypes, and the Enet algorithm was used to construct the NETosis-Related Riskscore (NETRS). Enrichment analyses were conducted using the GSVA and ClusterProfiler packages. Six distinct algorithms were utilized to evaluate patients' immune cell infiltration level. Patients' SNV and CNV data were analyzed by maftools and GISTIC2.0, respectively. Drug information was obtained from the GDSC1, and predicted by the Oncopredict package. Patient response to immunotherapy was evaluated by the TIDE algorithm in conjunction with the phs000452 immunotherapy cohort. Six NRGs' differential expression was verified using qRT-PCR and immunohistochemistry. Results: Among all cell types, neutrophils had the highest AUCell score. By Intersecting the DEGs between high and low NETosis classes, DEGs between normal and LUAD tissues, and prognostic related genes, 61 prognostic related NRGs were identified. Based on the 61 NRGs, all LUAD patients can be divided into two clusters, showing different prognostic and TME characteristics. Enet regression identified the NETRS composed of 18 NRGs. NETRS significantly associated with LUAD patients' clinical characteristics, and patients at different NETRS groups showed significant differences on prognosis, TME characteristics, immune-related molecules' expression levels, gene mutation frequencies, response to immunotherapy, and drug sensitivity. Besides, NETRS was more powerful than 20 published gene signatures in predicting LUAD patients' survival. Nine independent cohorts confirmed that NETRS is also valuable in predicting the prognosis of all NSCLC patients. Finally, six NRGs' expression was confirmed using three independent datasets, qRT-PCR and immunohistochemistry. Conclusion: NETRS can serves as a valuable prognostic indicator for patients with NSCLC, providing insights into the tumor microenvironment and predicting the response to cancer therapy.

Machine-learning and combined analysis of single-cell and bulk-RNA sequencing identified a DC gene signature to predict prognosis and immunotherapy response for patients with lung adenocarcinoma.

BACKGROUND: Innate immune effectors, dendritic cells (DCs), influence cancer prognosis and immunotherapy significantly. As such, dendritic cells are important in killing tumors and influencing tumor microenvironment, whereas their roles in lung adenocarcinoma (LUAD) are largely unknown. METHODS: In this study, 1658 LUAD patients from different cohorts were included. In addition, 724 cancer patients who received immunotherapy were also included. To identify DC marker genes in LUAD, we used single-cell RNAsequencing data for analysis and determined 83 genes as DC marker genes. Following that, integrative machine learning procedure was developed to construct a signature for DC marker genes. RESULTS: Using TCGA bulk-RNA sequencing data as the training set, we developed a signature consisting of seven genes and classified patients by their risk status. Another six independent cohorts demonstrated the signature' s prognostic power, and multivariate analysis demonstrated it was an independent prognostic factor. LUAD patients in the high-risk group displayed more advanced features, discriminatory immune-cell infiltrations and immunosuppressive states. Cell-cell communication analysis indicates that tumor cells with lower risk scores communicate more actively with the tumor microenvironment. Eight independent immunotherapy cohorts revealed that patients with low-risk had better immunotherapy responses. Drug sensitivity analysis indicated that targeted therapy agents exhibited greater sensitivity to low-risk patients, while chemotherapy agents displayed greater sensitivity to high-risk patients. In vitro experiments confirmed that CTSH is a novel protective factor for LUAD. CONCLUSIONS: An unique signature based on DC marker genes that is highly predictive of LUAD patients' prognosis and response to immunotherapy. CTSH is a new biomarker for LUAD.

A cooperative network at the nuclear envelope counteracts LINC-mediated forces during oogenesis in C. elegans.

Oogenesis involves transduction of mechanical forces from the cytoskeleton to the nuclear envelope (NE). In Caenorhabditis elegans, oocyte nuclei lacking the single lamin protein LMN-1 are vulnerable to collapse under forces mediated through LINC (linker of nucleoskeleton and cytoskeleton) complexes. Here, we use cytological analysis and in vivo imaging to investigate the balance of forces that drive this collapse and protect oocyte nuclei. We also use a mechano-node-pore sensing device to directly measure the effect of genetic mutations on oocyte nuclear stiffness. We find that nuclear collapse is not a consequence of apoptosis. It is promoted by dynein, which induces polarization of a LINC complex composed of Sad1 and UNC-84 homology 1 (SUN-1) and ZYGote defective 12 (ZYG-12). Lamins contribute to oocyte nuclear stiffness and cooperate with other inner nuclear membrane proteins to distribute LINC complexes and protect nuclei from collapse. We speculate that a similar network may protect oocyte integrity during extended oocyte arrest in mammals.

Central Aortic Blood Pressure Waveform Estimation with a Temporal Convolutional Network.

A novel temporal convolutional network (TCN) model is utilized to reconstruct the central aortic blood pressure (aBP) waveform from the radial blood pressure waveform. The method does not need manual feature extraction as traditional transfer function approaches. The data acquired by the SphygmoCor CVMS device in 1,032 participants as a measured database and a public database of 4,374 virtual healthy subjects were used to compare the accuracy and computational cost of the TCN model with the published convolutional neural network and bi-directional long short-term memory (CNN-BiLSTM) model. The TCN model was compared with CNN-BiLSTM in the root mean square error (RMSE). The TCN model generally outperformed the existing CNN-BiLSTM model in terms of accuracy and computational cost. For the measured and public databases, the RMSE of the waveform using the TCN model was 0.55 ± 0.40 mmHg and 0.84 ± 0.29 mmHg, respectively. The training time of the TCN model was 9.63 min and 25.51 min for the entire training set; the average test time was around 1.79 ms and 8.58 ms per test pulse signal from the measured and public databases, respectively. The TCN model is accurate and fast for processing long input signals, and provides a novel method for measuring the aBP waveform. This method may contribute to the early monitoring and prevention of cardiovascular disease.

A deep learning network based on CNN and sliding window LSTM for spike sorting.

Spike sorting plays an essential role to obtain electrophysiological activity of single neuron in the fields of neural signal decoding. With the development of electrode array, large numbers of spikes are recorded simultaneously, which rises the need for accurate automatic and generalization algorithms. Hence, this paper proposes a spike sorting model with convolutional neural network (CNN) and a spike classification model with combination of CNN and Long-Short Term Memory (LSTM). The recall rate of our detector could reach 94.40% in low noise level dataset. Although the recall declined with the increasing noise level, our model still presented higher feasibility and better robustness than other models. In addition, the results of our classification model presented an accuracy of greater than 99% in simulated data and an average accuracy of about 95% in experimental data, suggesting our classifier outperforms the current "WMsorting" and other deep learning models. Moreover, the performance of our whole algorithm was evaluated through simulated data and the results shows that the accuracy of spike sorting reached about 97%. It is noteworthy to say that, this proposed algorithm could be used to achieve accurate and robust automated spike detection and spike classification.

HOTAIR knockdown impairs metastasis of cervical cancer cells by down-regulating metastasis-related genes.

This study investigated the role of LncRNA HOTAIR knockdown in the biological impacts on cervical cancer cells. The HOTAIR gene in two human cervical cancer cell lines was silenced with small interfering (si) RNA siHOTAIR. Proliferation, apoptosis, migration and invasion of cells were assessed following the knockdown. The expressions of Notch1, EpCAM, E-cadherin, vimentin and STAT3 were assessed using qRT-PCR and Western blotting analysis. Compared with controls, HOTAIR levels were reduced significantly, the OD values of cells were significantly decreased in proliferation assays, cell apoptosis was significantly increased, cell migration and invasion were significantly reduced after HOTAIR knockdown. Molecular analysis showed that Notch1, EpCAM, vimentin and STAT3 expressions were decreased significantly, while the expression of E-cadherin was significantly increased after HOTAIR knockdown. Rescue experiments further confirmed that Notch1 and STAT3 were involved in siHOTAIR-mediated reduction of migration and invasion of cervical cancer cells.IMPACT STATEMENTWhat is already known on this subject? Long non-coding RNAs including HOTAIR, is implicated in occurrence and development of cancer and have been explored to develop new therapeutic options for cancer.What do the results of this study add? HOTAIR silencing significantly reduces the viability and migration ability of cells and induces cell apoptosis, adding experimental data supporting the potential use of HOTAIR specific-siRNA as a therapeutic avenue for the cancer.What are the implications of these findings for clinical practice and/or further research? The finding from this study would help develop clinically applicable therapeutic avenues for the cancer and identify new treatment targets in the relevant pathways leading to new drugs or treatments.

MJL-1 is a nuclear envelope protein required for homologous chromosome pairing and regulation of synapsis during meiosis in C. elegans.

Interactions between chromosomes and LINC (linker of nucleoskeleton and cytoskeleton) complexes in the nuclear envelope (NE) promote homolog pairing and synapsis during meiosis. By tethering chromosomes to cytoskeletal motors, these connections lead to processive chromosome movements along the NE. This activity is usually mediated by telomeres, but in the nematode Caenorhabditis elegans, special chromosome regions called "pairing centers" (PCs) have acquired this meiotic function. Here, we identify a previously uncharacterized meiosis-specific NE protein, MJL-1 (MAJIN-Like-1), that is essential for interactions between PCs and LINC complexes in C. elegans. Mutations in MJL-1 eliminate active chromosome movements during meiosis, resulting in nonhomologous synapsis and impaired homolog pairing. Fission yeast and mice also require NE proteins to connect chromosomes to LINC complexes. Extensive similarities in the molecular architecture of meiotic chromosome-NE attachments across eukaryotes suggest a common origin and/or functions of this architecture during meiosis.

Recruitment of Polo-like kinase couples synapsis to meiotic progression via inactivation of CHK-2.

Meiotic chromosome segregation relies on synapsis and crossover (CO) recombination between homologous chromosomes. These processes require multiple steps that are coordinated by the meiotic cell cycle and monitored by surveillance mechanisms. In diverse species, failures in chromosome synapsis can trigger a cell cycle delay and/or lead to apoptosis. How this key step in 'homolog engagement' is sensed and transduced by meiotic cells is unknown. Here we report that in C. elegans, recruitment of the Polo-like kinase PLK-2 to the synaptonemal complex triggers phosphorylation and inactivation of CHK-2, an early meiotic kinase required for pairing, synapsis, and double-strand break (DSB) induction. Inactivation of CHK-2 terminates DSB formation and enables CO designation and cell cycle progression. These findings illuminate how meiotic cells ensure CO formation and accurate chromosome segregation.

Most animals, plants, and fungi reproduce sexually, meaning that the genetic information from two parents combines during fertilization to produce offspring. This parental genetic information is carried within the reproductive cells in the form of chromosomes. Reproductive cells in the ovaries or testes first multiply through normal cell division, but then go through a unique type of cell division called meiosis. During meiosis, pairs of chromosomes ­ the two copies inherited from each parent ­ must find each other and physically line up from one end to the other. As they align side-by-side with their partners, chromosomes also go through a mixing process called recombination, during which regions of one chromosome cross over to the paired chromosome to exchange information. Scientists are still working to understand how this process of chromosome alignment and crossing-over is controlled. If chromosomes fail to line up or cross over during meiosis, eggs or sperm can end up with too many or too few chromosomes. If these faulty reproductive cells combine during fertilization this can lead to birth defects and developmental problems. To minimize this problem, reproductive cells have a quality control mechanism during meiosis called "crossover assurance", which limits how often mistakes occur. Zhang et al. have investigated how cells can tell if their chromosomes have accomplished this as they undergo meiosis. They looked at egg cells of the roundworm C. elegans, whose meiotic processes are similar to those in humans. In C. elegans, a protein called CHK-2 regulates many of the early events during meiosis. During successful meiosis, CHK-2 is active for only a short amount of time. But if there are problems during recombination, CHK-2 stays active for longer and prevents the cell division from proceeding. Zhang et al. uncovered another protein that affects for how long CHK-2 stays switched on. When chromosomes align with their partners, a protein called PLK-2 sticks to other proteins at the interface between the aligned chromosomes. A combination of microscopy and test tube experiments showed that when PLK-2 is bound to this specific location, it can turn off CHK-2. However, if the chromosome alignment fails, PLK-2 is not activated to switch off CHK-2. Therefore, CHK-2 is only switched off when the chromosomes are properly aligned and move on to the next step in crossing-over, which then allows meiosis to proceed. Thus, PLK-2 and CHK-2 work together to detect errors and to slow down meiosis if necessary. Further experiments in mammalian reproductive cells will reveal how similar the crossover assurance mechanism is in different organisms. In the future, improved understanding of quality control during meiosis may eventually lead to improvements in assisted reproduction.

Performance Analysis of Electromyogram Signal Compression Sampling in a Wireless Body Area Network.

The rapid growth in demand for portable and intelligent hardware has caused tremendous pressure on signal sampling, transfer, and storage resources. As an emerging signal acquisition technology, compressed sensing (CS) has promising application prospects in low-cost wireless sensor networks. To achieve reduced energy consumption and maintain a longer acquisition duration for high sample rate electromyogram (EMG) signals, this paper comprehensively analyzes the compressed sensing method using EMG. A fair comparison is carried out on the performances of 52 ordinary wavelet sparse bases and five widely applied reconstruction algorithms at different compression levels. The experimental results show that the db2 wavelet basis can sparse EMG signals so that the compressed EMG signals are reconstructed properly, thanks to its low percentage root mean square distortion (PRD) values at most compression ratios. In addition, the basis pursuit (BP) reconstruction algorithm can provide a more efficient reconstruction process and better reconstruction performance by comparison. The experiment records and comparative analysis screen out the suitable sparse bases and reconstruction algorithms for EMG signals, acting as prior experiments for further practical applications and also a benchmark for future academic research.

Human-computer interaction based health diagnostics using ResNet34 for tongue image classification.

BACKGROUND AND OBJECTIVE: Tongue diagnosis is one of the characteristics of traditional Chinese medicine (TCM), but traditional tongue diagnosis is affected by many factors, and its differential diagnosis results are not widely recognized. The appearance of tongue diagnosis instruments is the product of the modernization of tongue diagnosis, and it has standard and objective advantages in clinical practice. In this study, based on standard tongue images, a tongue image dataset and detection model were constructed. And based on the deep learning convolutional neural network (CNN) algorithm and visual question answering technology, a human-computer interaction intelligent health detector for tongue image recognition is constructed. METHODS: In this research, 1420 tongue images were collected. After screening, experts judged them, and annotated the tongue images to form tongue image datasets. Then the artificial intelligence network framework based on deep learning convolutional neural network (CNN), that is, ResNet34, is applied to this dataset to automatically extract image features and realize tongue images classification. Finally, the VGG16 network framework is applied to the dataset to compare the classification model and compare with the classification effect. RESULTS: In this paper, relevant datasets were formed by collating the tongue images collected by annotation, which verified that the ResNet34 architecture could better perform the task of tooth mark and tongue feature recognition. Compared with similar learning tasks in existing studies, the accuracy of the teeth-printed tongue recognition model proposed in this study is more than 10% higher, which indicates that the CNN algorithm can distinguish teeth-printed tongue more accurately and effectively. At the same time, using datasets and models combined with visual question and answer technology, an AI health detector for TCM tongue image identification is designed, which can make health assessments and give suggestions to users. CONCLUSION: This study adopts a convolutional neural network model based on deep learning, which can reduce the extraction of tongue features more quickly and conveniently. At the same time, the model architecture has excellent performance and strong generalization ability and is more accurate in judging users' health status.

Formation characteristics of disinfection byproducts from four different algal organic matter during chlorination and chloramination.

Algal organic matter (AOM) has become an important precursor of disinfection byproducts (DBPs) in multiple drinking water sources. In this study, the formation of DBPs during chlorination and chloramination of AOMs from four algal species (Microcystis aeruginosa, Chlorella vulgaris, Scenedesmus obliquus, and Cyclotella sp.) under different conditions (disinfectant doses 4.0-8.0 mg/L as Cl2, pH 6.0-8.0, and bromide 0-1.0 mg/L) were simultaneously investigated. Some common and specific characteristics of DBP formation have also been identified. The yields of total DBPs from the four AOMs were 3.28 × 102-6.00 × 102 and 1.97 × 102-3.70 × 102 nmol/mg C during chlorination and chloramination, respectively. The proportions of haloacetic acids (HAAs) in total DBPs were approximately ≥50%. Increasing disinfectant doses or pH only enhanced the yields of trihalomethanes (THMs) during chlorination but enhanced the yields of THMs, HAAs and dihaloacetonitriles (DHANs) during chloramination. Increasing bromide concentrations enhanced THM yields but decreased HAA yields during chlorination and chloramination, in addition to the shift from chlorinated DBPs to brominated DBPs. The DHAN yields of the four AOMs slightly decreased with bromide levels during chlorination, whereas different AOMs showed different trends with bromide levels during chloramination. During chlorination, C. vulgaris and S. obliquus AOMs generated higher THM and DHAN yields (at 4.0-5.0 mg/L as Cl2) than the other AOMs. During chloramination, M. aeruginosa AOM generated higher THM and HAA yields than the other AOMs (at 0.1 mg/L bromide). Cyclotella sp. AOM had the highest THM-bromine substitution factors during chlorination and the highest DHAN-bromine substitution factors during both chlorination and chloramination (at 0.1 mg/L bromide).

Presence of bromide and iodide promotes the horizontal transfer of antibiotic resistance genes during chlorination: A preliminary study.

Chlorination was reported to have a great potential to increase horizontal gene transfer (HGT) of antibiotic resistance genes (ARGs), which poses a great threat to global human health. Bromide (Br-) and iodide (I-) ions are widely spread ions in water and wastewater. In chlorination, Br- and I- can be oxidized to active bromine and iodine species. The influence of the co-existing different halogen oxidants (chlorine + bromine or iodine species) on HGT of ARGs were rarely investigated. In this study, the conjugative transfer of ARGs between a donor strain E. coli K12 and a recipient strain E. coli HB101 was investigated in chlorination without/with the presence of Br- or I-. Immediately after the addition of sodium hypochlorite, 53-88 % of the dosed chlorine was rapidly consumed, 10 %-42 % fast transformed into organic combined chloramines, and only low levels of free chlorine (0.02-0.8 mg/L as Cl2) left in the diluted cultural medium. Conjugative transfer mediated by the RP4 plasmid was not significantly enhanced in chlorination without the presence of Br- or I-. With the presence of Br- (0.5-5.0 mg/L) or I- (0.05-0.5 mg/L) in chlorination, the co-existing free halogen oxidants and their organic combined ones up-regulated the mRNA expression of the oxidative stress-regulatory gene (rpoS), outer membrane protein gene (ompC), and conjugation-relevant genes (trbBp and trfAp), and caused more damage to cell entirety. As a result, the co-existing reactive halogen oxidants enhanced the HGT of ARGs probably via conjugative transfer and transformation. This study showed that the presence of Br- and I- of common levels in aquatic environment promoted HGT of ARGs in chlorination, thus accelerating the transmission and prevalence of ARGs.