A study on predicting the length of hospital stay for Chinese patients with ischemic stroke based on the XGBoost algorithm

BACKGROUND

The incidence of stroke is a challenge in China, as stroke imposes a heavy burden on families, national health services, social services, and the economy. The length of hospital stay (LOS) is an essential indicator of utilization of medical services and is usually used to assess the efficiency of hospital management and patient quality of care. This study established a prediction model based on a machine learning algorithm to predict ischemic stroke patients' LOS.

METHODS

A total of 18,195 ischemic stroke patients' electronic medical records and 28 attributes were extracted from electronic medical records in a large comprehensive hospital in China. The prediction of LOS was regarded as a multi classification problem, and LOS was divided into three categories: 1-7 days, 8-14 days and more than 14 days. After preprocessing the data and feature selection, the XGBoost algorithm was used to build a machine learning model. Ten fold cross-validation was used for model validation. The accuracy (ACC), recall rate (RE) and F1 measure were used to evaluate the performance of the prediction model of LOS of ischemic stroke patients. Finally, the XGBoost algorithm was used to identify and remove irrelevant features by ranking all attributes based on feature importance.

RESULTS

Compared with the naive Bayesian algorithm, logistic region algorithm, decision tree classifier algorithm and ADaBoost classifier algorithm, the XGBoot algorithm has higher ACC, RE and F1 measure. The average ACC, RE and F1 measure were 0.89, 0.89 and 0.89 under the 10-fold cross-validation. According to the analysis of the importance of features, the LOS of ischemic stroke patients was affected by demographic characteristics, past medical history, admission examination features, and operation characteristics. Finally, the features in terms of hemiplegia aphasia, MRS, NIHSS, TIA, Operation or not, coma index etc. were found to be the top features in importance in predicting the LOS of ischemic stroke patients.

CONCLUSIONS

The XGBoost algorithm was an appropriate machine learning method for predicting the LOS of patients with ischemic stroke. Based on the prediction model, an intelligent medical management prediction system could be developed to predict the LOS based on ischemic stroke patients' electronic medical records.

Non-microbial methane emissions from tropical rainforest soils under different conditions

Non-microbial methane (NM-CH4), emissions from soil might play a significant role in carbon cycling and global climate change. However, the production mechanisms and emission potential of soil NM-CH4 from tropical rainforest remain highly uncertain. In order to explore the laws and characteristics of NM-CH4 emission from tropical rainforest soils. Incubation experiments at different environmental conditions (temperatures, soil water contents, hydrogen peroxide) and for soils with different soil organic carbon (SOC) contents were conducted to investigate the NM-CH4 emission characteristics and its influence factors of soils (0-10cm) that collected from a tropical rainforest in Hainan, China. Incubation results illustrated that soil NM-CH4 release showed a linear increase with the incubation time in the first 24 hours at 70 °C, whereas the logarithmic curve increase was found in 192 h incubation. Soil NM-CH4 emission rates under aerobic condition were significantly higher than that of under anaerobic condition at first 24 h incubation. The increasing of temperature, suitable soil water contents (0-100%), and hydrogen peroxide significantly promoted soil NM-CH4 emission rates at the first 24 h incubation. However, excessive soil water contents (200%) inhibited soil NM-CH4 emissions. According to the curve simulated from the NM-CH4 emission rates and incubation time at 70 °C of aerobic condition, soil would no longer release NM-CH4 after 229 h incubation. The NM-CH4 emissions were positively corelated with SOC contents, and the average soil NM-CH4 emission potential was about 6.91 ug per gram organic carbon in the tropical mountain rainforest. This study revealed that soils in the tropical rainforest could produce NM-CH4 under certain environment conditions and it supported production mechanisms of thermal degradation and reactive oxygen species oxidation. Those results could provide a basic data for understanding the soil NM-CH4 production mechanisms and its potential in the tropical rainforest.

VISPR-online: a web-based interactive tool to visualize CRISPR screening experiments

BACKGROUND

VISPR is an interactive visualization and analysis framework for CRISPR screening experiments. However, it only supports the output of MAGeCK, and requires installation and manual configuration. Furthermore, VISPR is designed to run on a single computer, and data sharing between collaborators is challenging.

RESULTS

To make the tool easily accessible to the community, we present VISPR-online, a web-based general application allowing users to visualize, explore, and share CRISPR screening data online with a few simple steps. VISPR-online provides an exploration of screening results and visualization of read count changes. Apart from MAGeCK, VISPR-online supports two more popular CRISPR screening analysis tools: BAGEL and JACKS. It provides an interactive environment for exploring gene essentiality, viewing guide RNA (gRNA) locations, and allowing users to resume and share screening results.

CONCLUSIONS

VISPR-online allows users to visualize, explore and share CRISPR screening data online. It is freely available at http://vispr-online.weililab.org , while the source code is available at https://github.com/lemoncyb/VISPR-online .

Molecular epidemiological characteristics of echovirus 6 in mainland China: extensive circulation of genotype F from 2007 to 2018

Echovirus 6 (E6) is associated with various clinical diseases and is frequently detected in environmental sewage. Despite its high prevalence in humans and the environment, little is known about its molecular phylogeography in mainland China. In this study, 114 of 21,539 (0.53%) clinical specimens from hand, foot, and mouth disease (HFMD) cases collected between 2007 and 2018 were positive for E6. The complete VP1 sequences of 87 representative E6 strains, including 24 strains from this study, were used to investigate the evolutionary genetic characteristics and geographical spread of E6 strains. Phylogenetic analysis based on VP1 nucleotide sequence divergence showed that, globally, E6 strains can be grouped into six genotypes, designated A to F. Chinese E6 strains collected between 1988 and 2018 were found to belong to genotypes C, E, and F, with genotype F being predominant from 2007 to 2018. There was no significant difference in the geographical distribution of each genotype. The evolutionary rate of E6 was estimated to be 3.631 × 10^-3 substitutions site^-1 year^-1 (95% highest posterior density [HPD]: 3.2406 × 10^-3-4.031 × 10^-3 substitutions site^-1 year^-1) by Bayesian MCMC analysis. The most recent common ancestor of the E6 genotypes was traced back to 1863, whereas their common ancestor in China was traced back to around 1962. A small genetic shift was detected in the Chinese E6 population size in 2009 according to Bayesian skyline analysis, which indicated that there might have been an epidemic around that year.

Initial CT features of COVID-19 predicting clinical category

PURPOSE

To analyze the initial CT features of different clinical categories of COVID-19.

MATERIAL AND METHODS

A total of 86 patients with COVID-19 were analyzed, including the clinical, laboratory and imaging features. The following imaging features were analyzed, the lesion amount, location, density, lung nodule, halo sign, reversed-halo sign, distribution pattern, inner structures and changes of adjacent structures. Chi-square test, Fisher's exact test, or Mann-Whitney U test was used for the enumeration data. Binary logistic regression analysis was performed to draw a regression equation to estimate the likelihood of severe and critical category. The forward conditional method was employed for variable selection.

RESULTS

Significant statistical differences were found in age (p = 0.001) and sex (p = 0.028) between mild and moderate and severe and critical category. No significant difference was found in clinical symptoms and WBC count between the two groups. The majority of cases (91.8%) showed multifocal lesions. The presence of GGO was higher in severe and critical category than in the mild and moderate category. (57.8% vs.31.7%, p = 0.015). Lymphocyte count was important indicator for the severe and critical category.

CONCLUSION

The initial CT features of the different clinical category overlapped. Combining with laboratory test, especially the lymphocyte count, could help to predict the severity of COVID-19.

SUPPLEMENTARY INFORMATION

The online version contains supplementary material available at 10.1007/s42058-021-00056-4.

Serum factor(s) from lung adenocarcinoma patients regulates the molecular clock expression

Lung cancer is a leading cause of cancer-associated deaths worldwide. Lung cancer may lead to circadian disruption, which could contribute to the development of lung cancer. Recently, several studies using animal models indicated that tumors influence systemic circadian homeostasis in remote tissues. However, it is unclear whether carcinoma of the lungs influences remote circadian rhythm, whether this effect exists in humans, and whether signals from the tumor travel through the blood. In this study, we used a cell-based assay to determine whether serum from patients with lung adenocarcinoma could modulate the molecular clock. We found that the daily oscillation period of Bmal1 was significantly lengthened following treatment with serum from untreated lung adenocarcinoma patients. In addition, heat inactivation of this serum abolished the effect, suggesting that a heat-sensitive circulating factor(s) is present in the serum of untreated lung adenocarcinoma patients. Using real-time PCR, we also examined the mRNA abundance of Bmal1, Cry1, and Per1 in human osteosarcoma u2os cell line, HUVECs and A549 cell lines. The expression of Bmal1 was changed in A549 cells in the presence of sera from lung adenocarcinoma patients. Our study revealed a direct effect of serum from lung adenocarcinoma patients on the molecular clock.

Research on spatial characteristics of metropolis development using nighttime light data: NTL based spatial characteristics of Beijing

As the capital and one of the metropolises in China, Beijing has met with a number of serious so-called "urban diseases" in the process of rapid urbanization such as blind expansion of urban areas, explosion of population and the increase of urban heat island effect. To treat these "urban diseases" and make the metropolis develop healthful and sustainable in Beijing in the future, the spatial characteristics of metropolis developments in Beijing are explored in this paper. The urban built-up areas in Beijing are extracted using the DMSP-OLS nighttime light data from 1992 to 2013. The characteristics of the urban developments of Beijing are studied, including spatial and temporal scales of urban developments, urban barycenter of Beijing and its transfer trajectory, variations of urban spatial forms and the differences of urban internal developments. The results have shown that the built-up areas had been increasing and circling extending from the central urban areas to the outer spaces in the last 21 years. The built-up area had expanded by 878km2 in 1992-2013, and the built-up area in 2013 had expanded to three times comparing to that of 1992. The expanding area of the built-up area in the northeast is the largest. The expansion of the urban had mainly occurred in 1996-2007, and the expanded area had accounted for 92% of the total research period. During the whole research period, the urban barycenter of Beijing had moved 5000.71 meters towards Northeast 28° of its original place from Dongcheng District to Chaoyang District. The development level of each municipal district had been increasing year by year, and the development differences among the municipal districts had been gradually reduced; the spatial forms of Beijing had been alternately changed between extensive and intensive expansion. The results of this study can help to plan urban land use and people migration of Beijing.

Genome-wide characterization of the abscisic acid-, stress- and ripening-induced (ASR) gene family in wheat (Triticum aestivum L.)

BACKGROUND

Abscisic acid-, stress-, and ripening-induced (ASR) genes are a class of plant specific transcription factors (TFs), which play important roles in plant development, growth and abiotic stress responses. The wheat ASRs have not been described in genome-wide yet.

METHODS

We predicted the transmembrane regions and subcellular localization using the TMHMM server, and Plant-mPLoc server and CELLO v2.5, respectively. Then the phylogeny tree was built by MEGA7. The exon-intron structures, conserved motifs and TFs binding sites were analyzed by GSDS, MEME program and PlantRegMap, respectively.

RESULTS

In wheat, 33ASR genes were identified through a genome-wide survey and classified into six groups. Phylogenetic analyses revealed that the TaASR proteins in the same group tightly clustered together, compared with those from other species. Duplication analysis indicated that the TaASR gene family has expanded mainly through tandem and segmental duplication events. Similar gene structures and conserved protein motifs of TaASRs in wheat were identified in the same groups. ASR genes contained various TF binding cites associated with the stress responses in the promoter region. Gene expression was generally associated with the expected group-specific expression pattern in five tissues, including grain, leaf, root, spike and stem, indicating the broad conservation of ASR genes function during wheat evolution. The qRT-PCR analysis revealed that several ASRs were up-regulated in response to NaCl and PEG stress.

CONCLUSION

We identified ASR genes in wheat and found that gene duplication events are the main driving force for ASR gene evolution in wheat. The expression of wheat ASR genes was modulated in responses to multiple abiotic stresses, including drought/osmotic and salt stress. The results provided important information for further identifications of the functions of wheat ASR genes and candidate genes for high abiotic stress tolerant wheat breeding.

Fine mapping of the major anthracnose resistance QTL AnRGO5 in Capsicum chinense 'PBC932'

BACKGROUND

Colletotrichum species are the causal agents of anthracnose, a major disease affecting the yield and quality of pepper (Capsicum spp.). Colletotrichum scovillei is widespread in China, has strong pathogenicity and drug resistance, and causes anthracnose disease in pepper fruits that severely reduces production. Previously, an anti-anthracnose locus AnR_GO5 was mapped to the P5 chromosome on the basis of analyses of fruit at the green mature stage. The aim of this study was to narrow down the interval of this locus and identify the gene responsible for conferring resistance.

RESULTS

On the basis of results of re-sequencing of Capsicum chinense 'PBC932' and C. annuum '77013', we developed Kompetitive allele-specific PCR (KASPar) markers and insertion-deletion (InDel) markers linked to AnR_GO5 at the green mature fruit stage and used them to construct a genetic linkage map (42 markers, 24.4 cM in length). Using data obtained in phenotypic and genotypic analyses of BC₄S₁, BC₄S₂, and BC₄S₃ populations, AnR_GO5 was located between the markers P5in-2266-404 and P5in-2268-978 within a physical distance of 164 kb. This region contained five genes, including CA05g17730. CA05g17730 encodes 'R1C-3-like' putative late blight resistance protein homologs. The transcript level of CA05g17730 differed between 'PBC932' and '77013'. The structure of the CA05g17730 gene also differed between 'PBC932' and '77013'.

CONCLUSIONS

We narrowed down the QTL interval to a region containing five genes. These results will be useful for further research on the mechanisms of resistance to anthracnose, and for marker assisted selection for anthracnose-resistant capsicum lines.

Integrated fate assessment of aromatic amines in aerobic sewage treatment plants

The fate and exposure of chemicals in sewage treatment plants (STPs) are major considerations in risk assessment and environmental regulation. The biodegradability and removal of seven aromatic amines were systematically evaluated using a three-tiered integrated method: a standard ready biodegradability test, an aerobic sewage treatment simulation method, and model prediction. In tier 1, the seven aromatic amines were not readily biodegraded after 28 days. In adapted aerobic active sludge, 4-isopropyl aniline, 2,4-diaminotoluene, and 4-nitroaniline among them exhibited the degradation half-life time less than 20 h, the other four aromatic amines exhibited persistent with degradation half-life of > 60 h. In tier 2 of the aerobic sewage treatment simulation testing, 2,4-diaminotoluene, 4-nitroaniline, and 4-isopropylaniline demonstrated moderately to high overall removal. Hydraulic retention time (HRT) affects the removal with the optimum HRT was determined to be 12 h to 24. 2,6-Dimethyl aniline, 2-chloro-4-nitroaniline, 2,6-diethylaniline, and 3,4-dichloroaniline were not removed during the test, indicting these four aromatic amines will enter surface water and hence pose a potential risk to aquatic ecology. Considering the lack of an STP model in China for regulation purposes, in tier 3, we developed a Chinese STP (aerobic) (abbreviated as C-STP(O)) model that reflects a universal scenario for China to predict the fate. The predicted degradation, volatilization, and absorption showed a close relationship to the physicochemical properties of the chemicals, and had same tendency with tier 2 simulation test. The prediction showed that biodegradation rather than absorption or volatilization was the main removal process of aromatic amines in aerobic STP. With the combination of modified kinetics test with C-STP (O) model, the chemical fate can be more accurately predicted than using only the readily biodegradation result.

Evaluation of surface texture of dried Hami Jujube using optimized support vector machine based on visual features fusion

The surface texture of dried jujube fruits is a significant quality grading criterion. This paper introduced a novel visual feature fusion based on connected region density, texture features, and color features. The single-scale Two-Dimensional Discrete Wavelet Transform was used to perform single-scale decomposition and reconstruction of dried Hami jujube image before visual features extraction. The connected region density was extracted by the two different algorithms, whereas the texture features were extracted by Gray Level Co-occurrence Matrix and the color features were extracted by image processing algorithms. Based on selected features which obtained by correlation analysis of visual features, the accuracy rate of the optimized Support Vector Machine classification model was 96.67%. In comparing with Extreme Learning Machine classification model and other fusion methods, the optimized Support Vector Machine based on selected visual features fusion was better.

Differential contributions of NO3 -/NH4 + to nitrogen use in response to a variable inorganic nitrogen supply in plantlets of two Brassicaceae species in vitro

BACKGROUND

The primary sources of nitrogen for plants have been suggested to be nitrate (NO3 -) and ammonium (NH4 +). However, when both nitrate and ammonium are simultaneously available to plants, it is very difficult to differentially quantify NO3 -/NH4 + utilization in culture media or soil. Consequently, the contribution of NO3 -/NH4 + to total inorganic nitrogen assimilation cannot be determined.

RESULTS

We developed a method called the bidirectional stable nitrogen isotope tracer to differentially quantify the nitrate and ammonium utilization by Orychophragmus violaceus (Ov) and Brassica napus (Bn) plantlets in vitro. The utilization efficiency of nitrate was markedly lower than the utilization efficiency of ammonium for plantlets of both Ov and Bn. In both Ov and Bn, the proportion of NO3 -/NH4 + utilization did not show a linear relationship with inorganic nitrogen supply. The Ov plantlets assimilated more nitrate than the Bn plantlets at the lowest inorganic nitrogen concentration.

CONCLUSIONS

Quantifying the utilization of nitrate and ammonium can reveal the differences in nitrate and ammonium assimilation among plants at different inorganic nitrogen supply levels and provide an alternate way to conveniently optimize the supply of inorganic nitrogen in culture media.