Prediction of the risk of cytopenia in hospitalized HIV/AIDS patients using machine learning methods based on electronic medical records.

Background: Cytopenia is a frequent complication among HIV-infected patients who require hospitalization. It can have a negative impact on the treatment outcomes for these patients. However, by leveraging machine learning techniques and electronic medical records, a predictive model can be developed to evaluate the risk of cytopenia during hospitalization in HIV patients. Such a model is crucial for designing a more individualized and evidence-based treatment strategy for HIV patients. Method: The present study was conducted on HIV patients who were admitted to Guangxi Chest Hospital between June 2016 and October 2021. We extracted a total of 66 clinical features from the electronic medical records and employed them to train five machine learning prediction models (artificial neural network [ANN], adaptive boosting [AdaBoost], k-nearest neighbour [KNN] and support vector machine [SVM], decision tree [DT]). The models were tested using 20% of the data. The performance of the models was evaluated using indicators such as the area under the receiver operating characteristic curve (AUC). The best predictive models were interpreted using the shapley additive explanation (SHAP). Result: The ANN models have better predictive power. According to the SHAP interpretation of the ANN model, hypoproteinemia and cancer were the most important predictive features of cytopenia in HIV hospitalized patients. Meanwhile, the lower hemoglobin-to-RDW ratio (HGB/RDW), low-density lipoprotein cholesterol (LDL-C) levels, CD4+ T cell counts, and creatinine clearance (Ccr) levels increase the risk of cytopenia in HIV hospitalized patients. Conclusion: The present study constructed a risk prediction model for cytopenia in HIV patients during hospitalization with machine learning and electronic medical record information. The prediction model is important for the rational management of HIV hospitalized patients and the personalized treatment plan setting.

The influence of cognitive level on the guaranteed behavioral response of landless farmers in the context of rural revitalization-An empirical study based on partial least squares structural equation modeling.

With the continuous acceleration of urbanization and agricultural modernization in China, the trend of concentration of rural land transfer is irreversible. For landless farmers, the absence of land guaranteed function inevitably gives rise to the substitution effect of other guaranteed methods. And the subjective preferences exhibited by farmers in making guaranteed behavior decisions can be quantitatively described as guaranteed behavioral responses (GBRs). Based on the analytical framework of distributed cognitive theory, this paper adopts the validated factor analysis method of structural equation modeling to quantitatively study the cognitive basis and behavioral responses of landless farmers' guaranteed behavior by combining the survey data of rural households in typical rural areas of Wuhan urban area. The study shows that the GBRs of landless farmers are significantly influenced by the cognitive level. "Locality power," "cultural power," and "personal power" are the main, important, and effective cognitive levels that influence farmers' GBRs, respectively. Policy-based protection occupies a dominant position in the rural social guaranteed system, savings-based protection still plays an important function in rural areas, and market-based protection has greater development potential.

A Study of the Differences in Heavy Metal Distributions in Different Types of Farmland in a Mining Area.

To investigate the effects of different types of land use and soil depths on the distributions of heavy metals in the soil in mining areas, heavy metals in different soil layers of five types of agricultural land in the Tongguan gold mining area were studied. The results revealed that the land use type had a greater impact than soil layers on the distribution of heavy metals in the soil. Among the five types of agricultural land examined, the risk values were only exceeded for the heavy metals lead(Pb) and mercury(Hg) in the pepper field, indicating combined pollution of Pb and Hg. Furthermore, some of the heavy metals, such as Pb, zinc(Zn), cadmium(Cd) and Hg, were highly significantly and positively correlated with each other. The pepper field should be monitored to prevent pollution from other heavy metals.

Analysis on hydraulic characteristics of improved sandy soil with soft rock.

Hydraulic properties of sandy soil from the Mu Us sandy land of Shaanxi Province were analyzed by using SEM technology. Soil porosity, the water characteristic curve, and unsaturated hydraulic conductivity of aeolian sandy soil with added soft rock were analyzed, and fractal characteristics were established. Soil hydraulic properties revealed the effect of soft rock application on soil structure and hydraulic properties. Mass ratios of soft rock to aeolian sand were 1:5, 1:2, and 1:1. Results showed that the addition of soft rock can significantly increase the bulk density of sandy soil and reduce the total porosity and macroporosity. The mass fraction of water-stable aggregates greater than 0.25mm increases significantly, increasing the fractal dimension of soil pores; reducing the soil saturated water content and saturated hydraulic conductivity. SEM technology and pore fractal theory were used to predict the soil salinity curve and unsaturated hydraulic conductivity of the improved saline soil.

The mineralization characteristics of organic carbon and particle composition analysis in reconstructed soil with different proportions of soft rock and sand.

The organic carbon mineralization process reflects the release intensity of soil CO2. Therefore, the study of organic carbon mineralization and particle composition analysis of soft rock and sand compound soil can provide technical support and a theoretical basis for soil organic reconstruction (soil structure, materials and biological nutrition). Based on previous research, four treatments were selected: CK (soft rock:sand=0:1), C1 (soft rock:sand=1:5), C2 (soft rock:sand=1:2) and C3 (soft rock:sand=1:1), respectively. Specifically, we analyzed the organic carbon mineralization process and soil particle composition by lye absorption, laser granulometer, and scanning electron microscope. The results showed that there was no significant difference in organic carbon content between C1, C2 , and C3 treatments, but they were significantly higher than in the CK treatment (P < 0.05). The organic carbon mineralization rate in each treatment accords with a logarithmic function throughout the incubation period (P < 0.01), which can be divided into a rapid decline phase in days 1 to 11 followed by a steady decline phase in days 11 to 30. The cumulative mineralization on the 11th day reached 54.96%-74.44% of the total mineralization amount. At the end of the incubation, the cumulative mineralization and potential mineralizable organic carbon content of the C1, C2 and C3 treatments were significantly higher than those of the CK treatment. The cumulative mineralization rate was also the lowest in the C1 and C2 treatment. The turnover rate constant of soil organic carbon in each treatment was significantly lower than that of the CK treatment, and the residence time increased. With the increase of volume fraction of soft rock, the content of silt and clay particles increased gradually, the texture of soil changed from sandy soil to sandy loam, loam , and silty loam, respectively. With the increase of small particles, the structure of soil appear ed to collapse when the volume ratio of soft rock was 50%. A comprehensive mineralization index and scanning electron microscopy analysis, when the ratio of soft rock to sand volume was 1:5-1:2, this can effectively increase the accumulation of soil organic carbon. Then, the distribution of soil particles was more uniform, the soil structure was stable (not collapsed), and the mineralization level of unit organic carbon was lower. Our research results have practical significance for the large area popularization of soft rock and sand compound technology.

Effects of row direction and row spacing on maize leaf senescence.

To analyze three row orientations (south-north, east-west, southwestern 20°) and two row spacings ('65 + 65', '160 + 40'), we investigated the effect of row orientation and planting pattern on photosynthetic performance, physiological and biochemical indicators related to the aging of leaves. Results revealed that during maturity stage, in north-south and east-west, the initial fluorescence (Fo) at '65 + 65' were higher than those under'160 + 40'; the maximum quantum yield of PS2 photochemistry(ΦP0), basal quantum yield of non-photochemical processes in PS2(ΦN0)of the lower leaves and photosynthetic rate of the upper and ear leaves under'160 + 40'were higher than those under'65 + 65'. The polyphenoloxidase (POD) activities of leaves at different positions under '160 + 40' were higher than that under'65 + 65', while the malondialdehyde (MDA) content was lower. The photosynthesis rate, superoxide dismutase (SOD) and catalase (CAT) activity of leaves at different positions under southwestern 20° '160 + 40' were higher than others. Whilst MDA content '160 + 40' were lower. Therefore, in De Hui City, Jilin Province, southwestern 20° '160 + 40' delayed leaf senescence at the late stage of growth of maize, as well as the effect of increasing maize yield was most obvious.

Three-dimensional bio-printing.

Three-dimensional (3D) printing technology has been widely used in various manufacturing operations including automotive, defence and space industries. 3D printing has the advantages of personalization, flexibility and high resolution, and is therefore becoming increasingly visible in the high-tech fields. Three-dimensional bio-printing technology also holds promise for future use in medical applications. At present 3D bio-printing is mainly used for simulating and reconstructing some hard tissues or for preparing drug-delivery systems in the medical area. The fabrication of 3D structures with living cells and bioactive moieties spatially distributed throughout will be realisable. Fabrication of complex tissues and organs is still at the exploratory stage. This review summarize the development of 3D bio-printing and its potential in medical applications, as well as discussing the current challenges faced by 3D bio-printing.

PepA and ArgR do not regulate Cre recombination at the bacteriophage P1 loxP site.

In the lysogenic state, bacteriophage P1 is maintained as a low copy-number circular plasmid. Site-specific recombination at loxP by the phage-encoded Cre protein keeps P1 monomeric, thus helping to ensure stable plasmid inheritance. Two Escherichia coli DNA-binding proteins, PepA and ArgR, were recently reported to be necessary for maintenance or establishment of P1 lysogeny. PepA and ArgR bind to regulatory DNA sequences upstream of the ColE1 cer recombination site to regulate site-specific recombination by the XerCD recombinases. This recombination keeps ColE1 in a monomeric state and helps to ensure stable plasmid maintenance. It has been suggested that ArgR and PepA play a similar role in P1 maintenance, regulating Cre recombination by binding to DNA sequences upstream of loxP. Here, we show that ArgR does not bind to its proposed binding site upstream of loxP, and that Cre recombination at loxP in its natural P1 context is not affected by PepA and ArgR in vitro. When sequences upstream of loxP were mutated to allow ArgR binding, PepA and ArgR still had no effect on Cre recombination. Our results demonstrate that PepA requires specific DNA sequences for binding, and that PepA and ArgR have no direct role in Cre recombination at P1 loxP.

Mutagenesis of PepA suggests a new model for the Xer/cer synaptic complex.

PepA is an aminopeptidase and also functions as a DNA-binding protein in two unrelated systems in Escherichia coli: Xer site-specific recombination and transcriptional regulation of carAB. In these systems, PepA binds to and brings together distant segments of DNA to form interwrapped, nucleosome-like structures. Here we report the selection of PepA mutants that were unable to support efficient Xer recombination. These mutants were defective in DNA-binding and in transcriptional regulation of carAB, but had normal peptidase activity. The mutations define extended patches of basic residues on the surface of the N-terminal domain of PepA that flank a previously proposed DNA-binding groove in the C-terminal domain of PepA. Our results suggest that DNA passes through this C-terminal groove in the PepA hexamer, and is bound by N-terminal DNA-binding determinants at each end of the groove. Based on our data, we propose a new model for the Xer synaptic complex, in which two recombination sites are wrapped around a single hexamer of PepA, bringing the cross-over sites together for strand exchange by the Xer recombinases. In this model, PepA stabilizes negative plectonemic interwrapping between two segments of DNA by passing one segment through the C-terminal groove while the other is held in place in a loop over the groove.