Search

BackgroundTreatment of cancer with pyroptosis is an emerging strategy. Molecular subtypes based on pyroptosis-related genes(PRGs) seem to be considered more conducive to individualized therapy. It is meaningful to construct a pyroptosis molecular subtypes-related prognostic signature (PMSRPS) to predict the overall survival (OS) of patients with pancreatic adenocarcinoma(PAAD) and guide treatment.MethodsBased on the transcriptome data of 23 PRGs, consensus clustering was applied to divide the TCGA and GSE102238 combined cohort into three PRGclusters. Prognosis-related differentially expressed genes(DEGs) among PRGclusters were subjected to LASSO Cox regression analysis to determine a PMSRPS. External cohort and in vitro experiments were conducted to verify this PMSRPS. The CIBERSORT algorithm, the ESTIMATE algorithm and the Immunophenoscore (IPS) were used to analyze the infiltrating abundance of immune cells, the tumor microenvironment (TME), and the response to immunotherapy, respectively. Wilcoxon analysis was used to compare tumor mutational burden (TMB) and RNA stemness scores (RNAss) between groups. RT-qPCR and in vitro functional experiments were used for evaluating the expression and function of SFTA2.ResultsBased on three PRGclusters, 828 DEGs were obtained and a PMSRPS was subsequently constructed. In internal and external validation, patients in the high-risk group had significantly lower OS than those in the low-risk group and PMSRPS was confirmed to be an independent prognostic risk factor for patients with PAAD with good predictive performance. Immune cell infiltration abundance and TME scores indicate patients in the high-risk group have typical immunosuppressive microenvironment characteristics. Analysis of IPS suggests patients in the high-risk group responded better to novel immune checkpoint inhibitors (ICIs) than PD1/CTLA4. The high-risk group had higher TMB and RNAss. In addition, 10 potential small-molecule compounds were screened out. Finally, we found that the mRNA expression of SFTA2 gene with the highest risk coefficient in PMSRPS was significantly higher in PAAD than in paracancerous tissues, and knockdown of it significantly delayed the progression of PAAD.ConclusionsPMSRPS can well predict the prognosis, TME and immunotherapy response of patients with PAAD, identify potential drugs, and provide treatment guidance based on individual needs.

Given the development and abundance of social media, studying the stance of social media users is a challenging and pressing issue. Social media users express their stance by posting tweets and retweeting. Therefore, the homogeneous relationship between users and the heterogeneous relationship between users and tweets are relevant for the stance detection task. Recently, graph neural networks (GNNs) have developed rapidly and have been applied to social media research. In this paper, we crawl a large-scale dataset of the 2020 US presidential election and automatically label all users by manually tagged hashtags. Subsequently, we propose a bipartite graph neural network model, DoubleH, which aims to better utilize homogeneous and heterogeneous information in user stance detection tasks. Specifically, we first construct a bipartite graph based on posting and retweeting relations for two kinds of nodes, including users and tweets. We then iteratively update the node's representation by extracting and separately processing heterogeneous and homogeneous information in the node's neighbors. Finally, the representations of user nodes are used for user stance classification. Experimental results show that DoubleH outperforms the state-of-the-art methods on popular benchmarks. Further analysis illustrates the model's utilization of information and demonstrates stability and efficiency at different numbers of layers.

Increasingly, cyber aggression becomes the prevalent phenomenon that erodes the social media environment. However, due to subjective and expense, the traditional self-reporting questionnaire is hard to be employed in the current cyber area. In this study, we put forward the prediction model for cyber aggression based on the cutting-edge deep learning algorithm. Building on 320 active Weibo users' social media activities, we construct basic, dynamic, and content features. We elaborate cyber aggression on three dimensions: social exclusion, malicious humour, and guilt induction. We then build the prediction model combined with pretrained BERT model. The empirical evidence shows outperformance and supports a stronger prediction with the BERT model than traditional machine learning models without extra pretrained information. This study offers a solid theoretical model for cyber aggression prediction. Furthermore, this study contributes to cyber aggression behaviors' probing and social media platforms' organization., Comment: 11 pages, 2 tables

The plant-based refers to plant-based raw materials or products that are available as the source of protein and fat. Utilization and development of walnuts as a plant-based, resulting in a high-quality protein-rich walnut plant-based product: walnut protein powder and walnut peptides. Progress in research on the application of walnuts as a plant-based has been advanced, solving the problem of wasted resources and environmental pollution caused by the fact that walnut residue, a product of walnuts after oil extraction, is often thrown away as waste, or becomes animal feed or compost. This paper reviews and summarizes the research and reports on walnut plant-based at home and abroad, focusing on the application of walnut plant-based in the preparation process (enzymatic and fermentation methods) and the biological activity of the walnut protein and walnut peptide, to provide a theoretical basis for the further processing of walnuts as a walnut plant-based. It can make full use of walnut resources and play its nutritional and health care value, develop and build a series of walnut plant-based products, improve the competitiveness of walnut peptide products, turn them into treasure, and provide more powerful guidance for the development of food and medicine health industry in Yunnan.

Additional file 2: Figure S2. (A) and (C) train queue lasso region, (B) and (D) test queue lasso region, (E), train queue forest map shows 14 DEirlncrna pairs determined by Cox proportional risk regression in the stepwise method, and (F) test queue forest map shows 13 DEirlncrna pairs determined by Cox proportional risk regression in the stepwise method. Figure S3. (A) Train queue Risk Score for 213 patients with COAD. the maximum inflection point is the cut-off point obtained by the AIC. (B), Test queue Risk Score for 213 patients with colon cancer. the maximum inflection point is the cut-off point obtained by the AIC. (C) The comparison of the 3-year ROC curve of the train cohort with other common clinical features shows the superiority of risk score. (D) The comparison of the 3-year ROC curve of the test cohort with other common clinical features shows the superiority of risk score. (E) Analysis of time-dependent receiver operating characteristic (ROC) curve in train queue. (F) Time dependent ROC curve analysis of test queue. Figure S4: The prognostic model was validated in the train and test cohorts. (A) Overall survival (OS) of the train queue. (B) Total lifetime of test queue. (C) Risk score distribution in the train queue. (D) Risk score distribution in test queue. (E) Scatter diagram of survival status of train queue. (F) Scatter diagram of survival state of test queue.

Text classification is a primary task in natural language processing (NLP). Recently, graph neural networks (GNNs) have developed rapidly and been applied to text classification tasks. As a special kind of graph data, the tree has a simpler data structure and can provide rich hierarchical information for text classification. Inspired by the structural entropy, we construct the coding tree of the graph by minimizing the structural entropy and propose HINT, which aims to make full use of the hierarchical information contained in the text for the task of text classification. Specifically, we first establish a dependency parsing graph for each text. Then we designed a structural entropy minimization algorithm to decode the key information in the graph and convert each graph to its corresponding coding tree. Based on the hierarchical structure of the coding tree, the representation of the entire graph is obtained by updating the representation of non-leaf nodes in the coding tree layer by layer. Finally, we present the effectiveness of hierarchical information in text classification. Experimental results show that HINT outperforms the state-of-the-art methods on popular benchmarks while having a simple structure and few parameters., COLING 2022(Oral)

The “fission electron-collection” neutron detector (FECND) is a fission-based neutron detector which has simple structure and could be used as a passive detector. In this work, a FECND prototype is fabricated and studied with thermal neutrons from a reactor. Due to the low sensitivity of FECND, the fission coat with an diameter of 93 mm and a235U mass thickness of 1 mg/cm2 was electroplated on the coated electrode of the prototype to obtain a robust signal. The experimental results indicate that the sensitivity of the prototype to thermal neutrons is 2.8 × 10−18C·cm2 and 1.3 × 10−18C·cm2 without a high voltage when the electrodes gap is 30 mm and 15 mm respectively. The sensitivities for various applied voltages are also provided for the two gaps. The data show that the both sensitivities were increased to about 1.0 × 10−17C·cm2 after a high voltage of one hundred volts was applied, and a higher voltage up to 4 kV would not increase the sensitivity distinctly. The impacts of different electrode gaps and various applied voltages are discussed.

Using integration of Bayesian Assessment with Hamacher Operator, corrosion reliability of oil & gas pipelines is thoroughly studied in this paper. With the Fussell–Vesely method, 106 minimal cut sets (MCS) of the fault tree are worked out. Based on attributes & correlation levels of events, the middle events and key bottom events are divided into three categories. For the basic events affiliated to the combined events with “Strong” correlation, correction factor is set, meanwhile taking amplification/reduction factor into account. The failure probability of each basic event is calculated using Hamacher Operator considering correlation. Based on Boolean Algorithms, disjoint calculation is adopted, whereby the first and second order of reliability of the corrosion system are obtained. Respectively, adopting Beta Distribution Function (Pass–Fail Type) and Negative Log-gamma Distribution Function (Exponential Type) to fit the first & second order of system reliability, finally the first lower approximate limit and the second lower approximate limit are obtained. Integrating corrosion reliability study with failure correlation, this paper establishes a more solid & precise basis for quantitative risk assessment on oil & gas systems. With a specific case study, the model established in this paper is verified, and the confidence interval proves to be [95%, 100%].

A single crystal chemical vapor deposition (scCVD) diamond detector has been successfully employed for neutron measurements in the EAST (Experimental Advanced Superconducting Tokamak) plasmas. The scCVD diamond detector coated with a 5 μm 6LiF (95% 6Li enriched) layer was placed inside a polyethylene moderator to enhance the detection efficiency. The time-dependent neutron emission from deuteron plasmas during neutral beam injection (NBI) heating was obtained. The measured results are compared with that of fission chamber detectors, which always act as standard neutron flux monitors. The scCVD diamond detector exhibits good reliability, stability and the capability to withstand harsh radiation environments despite its low detection efficiency due to the small active volume.

Based on fragility and internal & external hazard, an overall reliability model can be established as a theoretical basis for quantitative risk assessment on oil and gas pipeline systems. In this paper, overall reliability is regarded as conditional probability, including probability of third-party hazard and pipeline fragility. Third parties refer to all individuals, organizations and mechanical tools that cause unexpected damages to pipelines when they are carrying out operations either relevant or irrelevant to pipelines. By analyzing the propagation, waveform and frequency spectrum of natural seismic waves, the acceleration waveform of an artificial seism is simulated. Also a finite element model of pipeline's seismic response is established. Various pipelines' reactions simulated by ANSYS software are analyzed. Then the fragility probability of pipelines under blasting seismic actions is determined by normal distribution model. The overall reliability model is established and quantitatively evaluates the risks of Petro-China Gang-Zao product oil pipeline. By comparing the calculation results from algorithm software with the actual situation, the confidence interval of this model is [95%, 100%]. This model can precisely calculate the risk probability of certain pipe sections and then decide which section should be preferentially maintained and protected, increasing the efficiency of pipeline management.

Because of their prominent advantages in strategic energy reserves and peacetime peak adjustment, underground salt cavern gas storages have drawn attention worldwide. Moreover, embedding wastes in the underground space of salt caverns complies with the global theme of ecological environmental protection. The creep properties of underground salt cavern gas storages directly impact the safety and stability of the long-term operation of gas storages. In this study, the physical model of the cavity in a selected gas storage was established based on the creep law and creep constitutive model of salt rocks. FLAC 3D software was adopted to simulate the rheological displacement and related plastic deformation of the surrounding rock under varying internal pressures. Safety indexes, such as the cavity volume shrinkage and safety factors of surrounding rocks, were calculated for different operating conditions. By controlling the cavity volume shrinkage, both the minimum and maximum operating internal pressure values were obtained. An optimized design scheme including optimized cavity shape and height-diameter ratio was proposed to ensure the long-term safe and stable operation of gas storages.

A 2" × 2" BC501A liquid scintillation detector with a gain stabilization system is developed and applied to neutron and γ-ray measurement on the EAST tokamak. Energy calibration of a liquid scintillator using a fast coincidence method is presented and compared with the Monte Carlo simulation. Determination of the proton light output function of the BC501A is presented. Results from dedicated experiments with an Am-Be neutron source, γ source and quasi-monoenergetic neutron beams, and from measurements on EAST tokamak are presented and discussed.

The “fission electron-collection” neutron detector (FECND) is a fission-based neutron detector with flat and fast responses, and is a promising choice for the diagnostics of fusion neutron flux and spectrum. Due to its low sensitivity, the detector output is weak and is disturbed easily in calibration experiments. The backgrounds in the calibration primarily come from scattered neutrons and the γ rays induced by neutron inelastic scattering. A symmetrical structure of FECND was proposed to suppress the output of γ rays, and its effectiveness was validated by Monte Carlo simulations. A method was presented to eliminate the impact of scattered neutrons by utilizing the solid angle effect and the energy difference between source neutrons and scattered neutrons. Through calibration experiments and analyses, a D-T sensitivity of (8.6 ± 0.2)×10−20 C·cm2 was obtained for the FECND.

Due to the special physical properties of shale gas reservoirs, it is necessary to adopt unconventional and standardized technologies for its surface engineering construction. In addition, the surface engineering design of shale gas reservoirs in China faces many difficulties, such as high uncertainty of the gathering and transportation scale, poor adaptability of pipe network and station layout, difficult matching of the process equipments, and boosting production at the late stage. In view of these problems, the surface engineering construction of shale gas reservoirs should follow the principles of “standardized design, modularized construction and skid mounted equipment”. In this paper, standardized surface engineering design technologies for shale gas reservoirs were developed with the “standardized well station layout, universal process, modular function zoning, skid mounted equipment selection, intensive site design, digitized production management” as the core, after literature analysis and technology exploration were carried out. Then its application background and surface technology route were discussed with a typical shale gas field in Sichuan–Chongqing area as an example. Its surface gathering system was designed in a standardized way, including standardized process, the modularized gathering and transportation station, serialized dehydration unit and intensive layout, and remarkable effects were achieved. A flexible, practical and reliable ground production system was built, and a series of standardized technology and modularized design were completed, including cluster well platform, set station, supporting projects. In this way, a system applicable to domestic shale gas surface engineering construction is developed.