Stepwise Multiple Linear Regression Method Research Articles

Machine learning-based QSAR modeling, molecular docking, dynamics simulation studies for cytotoxicity prediction in MDA-MB231 triple-negative breast cancer cell line

Triple-negative breast cancer (TNBC) is the most aggressive cancer type that tests negative for progesterone, estrogen, and HER2 protein receptors. Therefore, TNBC is unlikely to respond either to drugs that target HER2 receptors or to hormonal therapy drugs. In the present study, 2D-QSAR model was developed to predict the inhibitory concentration (IC50) of terpene derivatives against human breast cancer metastatic cell line MDA-MB231, using V-Life MDS v4.5, module. The model was developed using the forward stepwise multiple linear regression method, with a regression coefficient (r2) of 0.86, and a cross-validated r2 (q2) of 0.8448. Molecular descriptors namely electronegativity (Epsilon-3), carbon atoms separated through five bond distances (T_C_C_5), Sum of Electrotopological state indices of –CH2 group (SssCH2count), Dipole moment of coordinate-z (Zcomp Dipole), and Distance between highest positive and negative electrostatic potential on van-der Waals surface area (Most +ve & -ve Potential Distance) were found to be the most potent contributing chemical descriptors for cytotoxicity against TNBC MDA-MB231 cell line. Additionally, binding affinities and interaction patterns revealed that the proposed compounds exhibit good binding affinities and substantial stability with c-Met-and β-tubulin receptors, as assessed by docking studies. Molecular dynamics simulations (100 ns) and binding free energy calculations were also performed using the MMGBSA method. The pharmacokinetic and eADME/T analysis of predicted compounds were assessed through Discovery Studio software. These findings may be of immense importance in the optimization and development of dual inhibitory potent anti-cancer inhibitors against the MDA-MB231 TNBC cell line. Moreover, this novel QSAR based prediction model/method is implemented in R-package and software tool developed for virtual screening purposes and available online for download through the public repository GitHub.

Structure-Activity Relationship Studies on VEGFR2 Tyrosine Kinase Inhibitors for Identification of Potential Natural Anticancer Compounds.

Over-expression of Vascular Endothelial Growth Factor Receptors (VEGFRs) leads to the hyperactivation of oncogenes. For inhibition of this hyperactivation, the USA Food Drug Administration (FDA) has approved many drugs that show adverse effects, such as hypertension, hypothyroidism, etc. There is a need to discover potent natural compounds that show minimal side effects. In the present study, we have taken structurally diverse known VEGFR2 inhibitors to develop a Quantitative Structure-Activity Relationship (QSAR) model and used this model to predict the inhibitory activity of natural compounds for VEGFR2. The QSAR model was developed through the forward stepwise Multiple Linear Regression (MLR) method. A developed QSAR model was used to predict the inhibitory activity of natural compounds. Absorption, Distribution, Metabolism, Excretion, and Toxicity (ADMET) assessment and molecular docking studies were performed. The binding stability of the natural compounds with VEGFR2 was elucidated through Molecular Dynamics (MD) simulation. The developed QSAR model against VEGFR2 showed the regression coefficient of the training dataset (r2) as 0.81 and the external regression coefficient of the test dataset (r2 test) 0.71. Descriptors, viz., electro-topological state of potential hydrogen bonds (maxHBint2, nHBint6), atom types (minssNH), maximum topological distance matrix (SpMAD_Dt), and 2D autocorrelation (ATSC7v), have been identified. Using this model, 14 natural compounds have been selected that have shown inhibitory activity for VEGFR2, of which six natural compounds have been found to possess a strong binding affinity with VEGFR2. In MD simulation, four complexes have shown binding stability up to 50ns. The developed QSAR model has identified 5 conserved activity-inducing physiochemical properties, which have been found to be correlated with the anticancer activity of the nonidentical ligand molecules bound with the VEGFR2 kinase. Lavendustin_A, 3'-O-acetylhamaudol, and arctigenin have been obtained as possible lead natural compounds against the VEGFR2 kinase.

Predicting saturated and near-saturated hydraulic conductivity using artificial neural networks and multiple linear regression in calcareous soils.

Hydraulic conductivity (Kψ) is one of the most important soil properties that influences water and chemical movement within the soil and is a vital factor in various management practices, like drainage, irrigation, erosion control, and flood protection. Therefore, it is an essential component in soil monitoring and managerial practices. The importance of Kψ in soil-water relationship, difficulties for its measurement in the field, and its high variability led us to evaluate the potential of stepwise multiple linear regression (SMLR), and multilayer perceptron (MLPNNs) and radial-basis function (RBFNNs) neural networks approaches to predict Kψ at tensions of 15, 10, 5, and 0 cm (K15, K10, K5, and K0, respectively) using easily measurable attributes in calcareous soils. A total of 102 intact (by stainless steel rings) and composite (using spade from 0-20 cm depth) soil samples were collected from different land uses of Fars Province, Iran. The common physico-chemical attributes were determined by the common standard laboratory approaches. Additionally, the mentioned hydraulic attributes were measured using a tension-disc infiltrometer (with a 10 cm radius) in situ. Results revealed that the most of studied soil structure-related parameters (soil organic matter, soluble sodium, sodium adsorption ratio, mean weight diameter of aggregates, pH, and bulk density) are more correlated with K5 and K0 than particle-size distribution-related parameters (sand, silt, and standard deviation and geometric mean diameter of particles size). For K15 and K10, the opposite results were obtained. The applied approaches predicted K15, K10, K5, and K0 with determination coefficient of validation data (R2val) of 0.52 to 0.63 for SMLR; 0.71 to 0.82 for MLPNNs; and 0.58 to 0.78 for RBFNNs. In general, the capability of the applied methods for predicting Kψ at all the applied tensions was ranked as MLPNNs > RBFNNs > SMLR. Although the SMLR method provided easy to use pedotransfer functions for predicting Kψ in calcareous soils, the present study suggests using the MLPNNs approach due to its high capability for generating accurate predictions.

Aversion-related effects of kappa-opioid agonist U-50488 on neural activity and functional connectivity between amygdala, ventral tegmental area, prefrontal cortex, hippocampus, and nucleus accumbens

Introduction: Among the various receptor systems in the brain, the opioid receptors have been the subject of extensive research due to their integral role in pain modulation, reward processing, and emotional regulation. The kappa-opioid receptor (KOR) system, in particular, stands apart due to its unique contribution to stress response, aversive behaviors, and dysphoric states. This paper aims to provide an understanding of the neural activity underlying the aversion-associated effects of the KOR agonist U-50488.
 Materials and Methods: Rats underwent stereotaxic surgery to implant electrodes into the amygdala, ventral tegmental area, prefrontal cortex, hippocampus, and nucleus accumbens. The rats were subjected to conditioned place preference test to measure aversion to U-50488. After that, local field potential (LFP) recordings were made. LFP data were processed and analyzed using spectral and coherence analysis methods. A stepwise multiple linear regression was employed to identify the LFP features most significantly correlated with aversion to U-50488.Results: The administration of U-50488 resulted in significant changes in LFP signals across multiple brain regions. These changes were particularly notable in the theta, gamma, and delta bands of brain waves (p<0.05). Theta and gamma activities were especially sensitive to the effects of U-50488. Connectivity calculations revealed shifts in coherence between brain regions, particularly highlighting the amygdala's involvement. While changes were also observed in the ventral tegmental area, prefrontal cortex, hippocampus, and nucleus accumbens (p<0.05), they contributed less to aversion. Using the stepwise multiple linear regression method, we established a final model with the 3 most significant variables: (1) coherence between the amygdala and medial prefrontal cortex, (2) coherence between the amygdala and hippocampus, and (3) theta power in the amygdala.
 Conclusion: Overall, the data provided insights into how electrical neural activity mediates aversion in response to KOR activation. The results showed that the severity of aversion can be reasonably predicted (r = 0.72±0.02, p = 0.0099) using LFP band power and functional connectivity data. We concluded that the amygdala is a brain region that contributes the most to the KOR agonist-induced aversion.

Experimental determination and QSAR analysis of the rate constants for [formula omitted] reactions with aromatic micropollutants in water

S(IV)-based systems used for advanced oxidation processes (AOPs) have been constructed for the degradation of organic contaminants via oxysulfur radicals, including SO3•−, SO4•−, and SO5•−. Although SO5•− is proposed as an active species in AOPs processes, research on the reactivity of SO5•− has remained unclear. In this work, 53 target aromatic micropollutants (AMPs), including 13 phenols, 27 amines, and 13 PPCPs were selected to determine the second-order reaction rate constants for SO5•− using the competitive kinetics method, in which the kSO5•− values, observed at pH 4 ranged from (2.44 ± 0.00) × 105 M−1 s−1 to (4.41 ± 0.28) × 107 M−1 s−1. Quantitative structure-activity relationship (QSAR) models for the oxidation of AMPs by SO5•− were developed based on 40 kSO5•− values of amines and phenols, and their molecular descriptors, using the stepwise multiple linear regression method. This comprehensive model exhibited the excellent goodness-of-fit (Radj2 = 0.802), robustness (QLOO2 = 0.749), and predictability (Qext2 = 0.656), and the one-electron oxidation potential (Eox), energy of the highest occupied molecular orbital energy (EHOMO), and most positive net atomic charge on the carbon atoms (qC+) were considered the most influential descriptors for the comprehensive model, indicating that SO5•− oxidizes pollutants via single electron transfer reaction and exhibits a strong oxidation capacity, especially for pollutants containing electron-donating groups. Moreover, the kSO5•− values of 13 PPCPs were predicted using this comprehensive model, which suggested the practical application significance of the QSAR model. This study emphasizes the direct oxidation capacity of SO5•−, which is important to evaluate and simulate AOPs based on S(IV).

Solvatochromism in mixed-chelate of copper(II) with tetramethyl ethylenediamine and 2,4-dihydroxybenzophenone, experimental and computational study

A mixed-chelate copper(II) complex, [Cu(dhbp)(tmeda)(H2O)]ClO4, where dhbp = 2,4-dihydroxybenzophenone and tmeda = tetramethyl ethylenediamine, was synthesized and characterized using various techniques including elemental analysis, molar conductance, UV-Vis, IR spectroscopy, and computational study. The solvatochromic behavior of the complex was investigated by visible spectroscopy, revealing positive solvatochromism. To understand the interaction mechanism between the complex and solvent, six standard solvent parameters (DN, AN, α, β, π*, and ET(30)) were assigned using the stepwise multiple linear regression (SMLR) method. Statistical analysis indicated that the DN parameter of the solvent had the most significant contribution in changing the d-d absorption band of the complex. Density functional theory (DFT) and time-dependent density functional theory (TD-DFT) calculations were performed to study the electronic structure of the complex and gain insights into the solvatochromism processes. The results suggested that the approach of solvent molecules and coordination to the vacant site of the complex led to a change in its geometry from square pyramidal to octahedral. This geometric transformation caused the d-d transition band of the complex to shift towards the red region with increasing solvent DN, thus explaining the observed solvatochromism.

Drivers of global surface urban heat islands: Surface property, climate background, and 2D/3D urban morphologies

Local climate of the city is modulated by not only land surface properties and climate background but also urban morphology. While the modulations of the former two have been extensively studied, there is still a lack of understanding of the importance of urban morphology, especially the three-dimensional (3D) morphology at a global scale. Here we examine the surface urban heat island (SUHI) intensities in 688 urban agglomerations worldwide and explore their driving factors in terms of surface properties, climate background, and 2D/3D urban morphologies, by using stepwise multiple linear regression and interpretable machine learning method. The results show that the interpretable machine learning method performs better in modeling SUHIs than traditional linear regression. Notably, urban morphologies, particularly its 3D pattern, have significant impacts on SUHIs. The overall contribution of urban morphology is comparable to that of surface property, and even surpasses the latter during the daytime in autumn and winter. Specifically, increased urban patch size or higher urban density can significantly amplify the SUHI intensities, while urban expansion characterized by scattered or multi-centered development exhibits the potential in mitigating SUHIs. In terms of 3D urban morphology, building height emerges as a crucial factor in shaping SUHIs. A rougher urban surface characterized by high-rise buildings can yield a cooling effect during the daytime while exacerbating the warming effect at night. Overall, our research provides valuable insights into the role of urban morphology in shaping SUHIs, offering guidance for urban planning strategies that promote sustainable and resilient cities.

Prediction models and major controlling factors of antibiotics bioavailability in hyporheic zone.

Recently, antibiotics have been frequently detected in the hyporheic zone (HZ) as a novel contaminant. Bioavailability assessment has gradually attracted more attention in order to provide a more realistic assessment of human health risks. In this study, two typical antibiotics, oxytetracycline (OTC) and sulfamethoxazole (SMZ), were used as target pollutants in the HZ of the Zaohe-Weihe River, and the polar organics integrated sampler was used to analyze the variation of antibiotics bioavailability. According to the characteristics of the HZ, the total concentration of pollutants, pH, and dissolved oxygen (DO) were selected as major predictive factors to analyze their correlation with the antibiotics bioavailability. Then the predictive antibiotic bioavailability models were constructed by stepwise multiple linear regression method. The results showed that there was a highly significant negative correlation between OTC bioavailability and DO (P < 0.001), while SMZ bioavailability showed a highly significant negative correlation with total concentration of pollutants (P < 0.001) and a significant negative correlation with DO (P < 0.01). The results of correlation analysis were further verified by Principal Component Analysis. Based on the experimental data, we constructed eight prediction models for the bioavailability of two antibiotics and verified them. The data points of the six prediction models were distributed in the 95% prediction band, indicating that the models were more reliable and accurate. The prediction models in this study provide reference for the accurate ecological risk assessment of the bioavailability of pollutants in the HZ, and also provide a new idea for predicting the bioavailability of pollutants in practical applications.

Experimental Study on Rapid Determination Method of Coal Seam Gas Content by Indirect Method

In view of the problems of heavy work, long cycle, high cost and low efficiency in the process of indirect determination of coal seam gas content, the basic gas parameters and coal quality indexes of 24 coal samples from 5 coal mines in the Hancheng area of Shanxi Province are measured by the laboratory measurement method. The raw coal gas content–gas desorption index of drilling cuttings (W–K1) relationship model is characterized by logarithmic function. Using SPSS data analysis software, a stepwise multiple linear regression method is used for statistical analysis. The results show that the factors that have a significant impact on the regression slope C in the W–K1 relationship model are gas adsorption constant (a), apparent density (ARD), initial velocity of gas diffusion (Δp) and consistent coefficient (f). The factors that have a significant impact on the regression constant D are Δp and atmospheric adsorption (Q). Then, the mathematical model of rapid prediction of coal seam gas content is determined. Compared with the measured values, the average absolute error rate is 12.84%, which meets the prediction requirements and provides a simple and easy method for rapid determination of coal seam gas content in coal mines in the Hancheng area.

Quantitative structure-activity relationship for the photooxidation of aromatic micro-pollutants induced by graphene oxide in water

The photocatalytic degradation behavior of aromatic micro-pollutants (AMPs) exhibits complexity and uncertainty, which mainly depends on the properties of different substituents on benzene. And with similar catalytic reaction substrates, the reaction rate constant could reveal the influence of different characteristics of molecular structure in a specific system. Therefore, the photooxidation pseudo first-order kinetic rate constants (kobs) of 30 AMPs were experimentally determined in Photo-GO system. A quantitative structure-activity relationship (QSAR) model for predicting the photooxidation reaction of AMPs has been developed by stepwise multiple linear regression (MLR) method, based on the lg kobs and representative molecule descriptors (20 in total) including physicochemical, quantum chemical and electrostatic descriptors. Afterwards, Radj2, QLOO2, and Qext2 were calculated as 0.870, 0.841, and 0.732 respectively, which exhibited the excellent goodness-of-fit, robustness, and predictability of the QSAR model, indicating its great prediction ability for photooxidation behavior of AMPs. Meanwhile, during the photooxidation process of AMPs with GO, the model revealed that the one-electron oxidation potential (Eox), molecular dipole moment (μ), and number of hydrogen bond donors (#HD) were the most important molecular structural parameters, which showed that the single electron transfer pathway and adsorption were as the significant steps. Additionally, the Hammett correlation showed that photooxidation of AMPs in Photo-GO system is of typical electrophilic reactions, which demonstrated that the electron-donating substituents could promote the photooxidation of AMPs. The QSAR model was constructed and evaluated to perform the prediction of AMPs reaction kinetics, which provided a guidance for the study of the mechanism and selective oxidation of AOPs photooxidation system based on GO.

Predictive Models of Gas/Particulate Partition Coefficients (KP) for Polycyclic Aromatic Hydrocarbons and Their Oxygen/Nitrogen Derivatives.

Polycyclic aromatic hydrocarbons (PAHs) and their oxygen/nitrogen derivatives released into the atmosphere can alternate between a gas phase and a particulate phase, further affecting their environmental behavior and fate. The gas/particulate partition coefficient (KP) is generally used to characterize such partitioning equilibrium. In this study, the correlation between log KP of fifty PAH derivatives and their n-octanol/air partition coefficient (log KOA) was first analyzed, yielding a strong linear correlation (R2 = 0.801). Then, Gaussian 09 software was used to calculate quantum chemical descriptors of all chemicals at M062X/6-311+G (d,p) level. Both stepwise multiple linear regression (MLR) and support vector machine (SVM) methods were used to develop the quantitative structure-property relationship (QSPR) prediction models of log KP. They yield better statistical performance (R2 &gt; 0.847, RMSE &lt; 0.584) than the log KOA model. Simulation external validation and cross validation were further used to characterize the fitting performance, predictive ability, and robustness of the models. The mechanism analysis shows intermolecular dispersion interaction and hydrogen bonding as the main factors to dominate the distribution of PAH derivatives between the gas phase and particulate phase. The developed models can be used to predict log KP values of other PAH derivatives in the application domain, providing basic data for their ecological risk assessment.

Trophic state assessment of optically diverse lakes using Sentinel-3-derived trophic level index

• The TLI retrieval algorithm with a global OWT classification scheme can achieve higher accuracy. • Using S owt > 0.9 to group training data for each OWT can help build more robust algorithms. • The eutrophic of lakes in Wuhan haven’t decreased during the COVID-19 lockdown period. An accurate estimation of trophic state of lakes with satellite remote sensing is a challenge due to the optical complexity and variability associated with inland waters. Match-up data from 393 sampling stations that has concurrent Sentinel-3 OLCI images were acquired across Wuhan lakes. Trophic Level Index (TLI) algorithms were developed within a global Optical Water Type (OWT) classification system. The performance of algorithms with limited training data gathered by using spectral similarity of highest S owt was not improved compared with that on basis of no classification. In contrast, using spectral similarity of S owt > 0.9 rather than the highest S owt to group more training data with similar traits for each OWT can help build more robust algorithms, which performance is better than that on basis of no classification. Algorithm performance statistics of the test dataset for the stepwise multiple linear regression (SMLR) method were the following: Mean Absolute Error (MAE) = 5.56; Mean Absolute Percentage Error (MAPE) = 11.02 %; Root Mean Square Error (RMSE) = 7.24 and for the back propagation neural network on the basis of the Levenberg-Marquardt-Bayesian regularization algorithm (LMBR-BPNN) method MAE = 4.56; MAPE = 8.33 %; RMSE = 5.98. We detected 8 different OWTs (2,3,4,5,9,10,11,12) in Wuhan lakes and clear spatio-temporal patterns of the trophic state between 2018 and 2020.Our results revealed that the trophic state of Wuhan lakes did not decrease as expected during the COVID-19 lockdown period.

Economic burden of malaria in Burundian children: An evidence for disease management in resource limited settings

In Burundi, malaria is the leading cause of morbidity and mortality, especially in children under five. This results in high clinical burden of the disease; however, its economic burden remains unknown. The aim of this study was to estimate the economic burden associated with malaria in Burundi and explore the factors that can affect the costs. This study was a prospective incidence-based cost-of-illness study analysed from the societal perspective. The study included children aged under five years with malaria infection, who visited and received treatment at any of the two study health facilities in the period of November to December 2019. Data collection was done by reviewing medical and financial records and by interviewing caregivers of the patients. Micro-costing approach was used to estimate the economic costs of malaria per episode. The cost was presented in international dollars (Int$) for the year 2019. Stepwise multiple linear regression method was applied to examine the factors affecting the costs and to generate a cost model. Eighty-five children with the average age of 29 months were included in the study. Most of them (70.5%) were treated as in-patient. The out-patient visits costed Int$23.5, while the in-patient hospitalization costed Int$218.2. The types of medical services, health facilities, antimalarials used, and duration of fever before seeking appropriate medical care were found to affect the costs. The model indicates that up to Int$18.76 can be saved per malaria episode if treated early (in less than 2 days). This can save up to Int$24,257,748 per year at national level. This study demonstrated that malaria is associated with a considerable economic burden in Burundi. It will support decision makers in deciding an appropriate clinical management for malaria prevention like the community case management program.

Spatial-Temporal Driving Factors of Urban Landscape Changes in the Karst Mountainous Regions of Southwest China: A Case Study in Central Urban Area of Guiyang City

Rapid urban expansion has significantly altered the regional landscape pattern, posing a serious threat to the sustainable development of natural and social ecosystems. By using landscape patterns indices and an area transfer matrix, this study analyzed the spatial-temporal changes of landscape patterns in the karst mountainous cities of southwest China from 2000 to 2020, by taking the central urban area of Guiyang City (CUAG) as the study area. This study explored the spatial and temporal driving factors of landscape pattern changes by using stepwise multiple linear regression and geographic detector methods. The results show: (1) CUAG’s landscape types altered changed drastically, with the area of forestland and construction land rapid increment and cultivated land decrement significantly. (2) The patches of construction land and forestland tended to be aggregated, the degree of fragmentation was reduced, and the shape was complex; cultivated land fragmentation was intensified. The connectivity of the landscape was improved, while the level of landscape diversity declined, the trend of landscape homogenization was obvious. (3) Socioeconomic and geographical endowment drivers have determined landscape pattern changes. The findings of this study may be used to interpret other similar landscapes worldwide and may imply the protection of urban ecosystem and sustainable development.

Pedestrian Road Crossing Behavior and Optimal Selection of Pedestrian Facilities at Mid-Block Crossings

Pedestrians are considered the most vulnerable of all road users, particularly in developing counties. Therefore, the safety of pedestrians plays a significant role during road crossing. The pedestrian gap acceptance and level of service (LOS) can be used to evaluate pedestrian crossing behavior. The existing statistical models are failed to predict the accurate level of pedestrian safety due to various assumptions. Thus, the objectives of this study are to develop pedestrian crossing behavior models, level of service model, and systematic methodology to select optimal pedestrian crossing facilities at midblock under mixed traffic conditions. The required data were collected from selected six midblock locations in Tiruchirappalli, India, by conducting videographic and questionnaire surveys simultaneously. Various influencing parameters on pedestrian crossing behavior were identified by conducting statistical correlation analysis. First, the pedestrian field critical gap was estimated and compared with various methods such as Raff, Greenshield, Ashworth, and Highway Capacity Manual 2010. The pedestrian gap acceptance model was developed by adopting stepwise multiple linear regression and non-linear artificial neural network methods. The developed models were validated and compared using the field data. The pedestrian LOS model was developed by utilizing the conventional regression method by using qualitative and quantitative data. Pedestrian LOS model was validated using another set of field data. Finally, the approximate solution to find the best crossing facility was found by performing weighting by a ranking method and analytical hierarchy process in all six locations. This study helps in identifying the most significant factors influencing road crossing behavior of pedestrians for the selected locations. The study results help design pedestrian facility and thereby improve pedestrian safety at uncontrolled intersections.

Impact of land-use/land-cover and landscape pattern on seasonal in-stream water quality in small watersheds

Changes in land use land cover (LULC) and landscape pattern in arid areas can considerably impact water quality in small watersheds. Exploring the associations between land use, landscape pattern, and riverine water quality can inform land-use planning for water-quality security. In northwest China's Jing River Watershed (JRW), water samples were collected at 67 points in 2016–2017 and analyzed. Coupled with Sentinel-2 remote-sensing images, this study calculated the Canadian Water Quality Index (CWQI) to assign the water-quality category. The Random Forest classification characterized current LULC patterns and computed landscape metrics in a riparian buffer zone. The stepwise multiple linear regression and redundancy analysis methods investigated the relationships amongst LULC, eight landscape metrics, and eight water-quality parameters. The results indicated high concentrations of total phosphorus (TP) and total nitrogen (TN) in the JRW's middle and lower reaches. For CWQI, lacking evident temporal fluctuations was contrasted by noticeable spatial variations. Intense Human activities have aggravated the deterioration of the water quality of the JRW. LULC-type changes, including Waterbody, Forest-grassland, and Farmland, imposed notable impacts on seasonal water quality. For landscape metrics, patch cohesion index (COHESION) strongly correlated with chemical oxygen demand (COD), mineralization degree (MD), and TN. By contrast, patch density (PD) correlated negatively with TN. Interspersion juxtaposition index (IJI) strongly correlated with biochemical oxygen demand after five days (BOD5). COHESION, PD, and IJI can be critical factors in predicting water-quality changes at the small-watershed scale. Our findings offered a robust scientific basis to refine regional sustainable water development strategy and rational water-resource management.

Regionalization of hydrological model parameters using gradient boosting machine

Abstract. The regionalization of hydrological model parameters is key to hydrological predictions in ungauged basins. The commonly used multiple linear regression (MLR) method may not be applicable in complex and nonlinear relationships between model parameters and watershed properties. Moreover, most regionalization methods assume lumped parameters for each catchment without considering within-catchment heterogeneity. Here we incorporated the Penman–Monteith–Leuning (PML) equation into the Distributed Time Variant Gain Model (DTVGM) to improve the mechanistic representation of the evapotranspiration (ET) process. We calibrated six key model parameters, grid by grid across China, using a multivariable calibration strategy which incorporates spatiotemporal runoff and ET datasets (0.25∘; monthly) as reference. In addition, we used the gradient boosting machine (GBM), a machine learning technique, to portray the dependence of model parameters on soil and terrain attributes in four distinct climatic zones across China. We show that the modified DTVGM could reasonably estimate the runoff and ET over China using the calibrated parameters but performed better in humid rather than arid regions for the validation period. The regionalized parameters by the GBM method exhibited better spatial coherence relative to the calibrated grid-by-grid parameters. In addition, GBM outperformed the stepwise MLR method in both parameter regionalization and gridded runoff simulations at a national scale, though the improvement pertaining to watershed streamflow validation is not significant due to most of the watersheds being located in humid regions. We also revealed that the slope, saturated soil moisture content, and elevation are the most important explanatory variables to inform model parameters based on the GBM approach. The machine-learning-based regionalization approach provides an effective alternative to deriving hydrological model parameters from watershed properties, particularly in ungauged regions.

A Study on Students’ Public Stigma of COVID-19 and Its Influencing Factors in The Context of The Long Term Handling of The COVID-19 Pandemic

Objective: To understand the public stigma and associated factors of COVID-19 in Jingzhou adolescents relating to the prevention and containment of COVID-19 cases. Methods: From June to October 2021, an online survey was conducted in two junior high schools in Jingzhou City. A total of 2888 students participated in the survey, with 2800 being valid responses (effective rate 97.0%). The survey tools used the COVID-19 attitude questionnaire, and the related factors were analyzed by a stepwise multiple linear regression method. Results: 31.4% of the participants showed public stigma towards COVID-19, with a larger portion of those being in lower grades (P&lt;0.01), only children (P&lt;0.05), and those not quarantined during the outbreak (P&lt;0.05). Multiple linear regression analyses showed higher levels of stress (P&lt;0.01), lower levels of social support (P&lt;0.05), younger age (P&lt;0.01) and higher educational level of parents (P&lt;0.05) during the pandemic correlated to stronger public stigma towards COVID-19. Conclusion: Participants' stigma of COVID-19 is widespread, and it is manifested in social isolation. The respondents’ level of stress and social support during the pandemic is shown to be linked to their degree of stigma. Methods for reducing this public stigma for COVID-19 may include stopping the spread of misinformation, providing psychological support, and alleviating negative emotions. Keywords: COVID-19, Adolescent, Student, Public Stigma.

Nondestructive evaluation of Chinese cabbage quality using mechanical vibration response

Mechanical vibration response using a self-made hammering device was performed to non-destructively measure Chinese cabbage quality using multidirectional vibration indices. To obtain a wide range of texture and different chemical attributes in cabbages, experiments were conducted every 3 d over 16 d of storage. The vibration parameters of each cabbage were extracted from the optimized frequency-response curves in both X-axis and Z-axis directions. A total of 26 features were extracted from the time and frequency domain signals. Cabbage quality attributes were determined by compactness and chemical indicators (soluble solids content, moisture content, Vitamin C content, and crude fiber content), and the newly introduced indicator of compactness was determined by a self-made apparatus. Stepwise multiple linear regression methods were used to quantitatively and qualitatively analyze the cabbage quality. Results showed that the compactness indicator was applicable to evaluate the texture of Chinese cabbage. The method which utilized mechanical vibration response was capable of evaluating and predicting cabbage quality indicators except for soluble solids content, particularly for compactness (rp = 0.757, RMSEP = 0.002 MPa). The better detection method for the vibration test was to stimulate the point in the equatorial part of a Chinese cabbage, and then collected the response signals at the point 180° apart from the excitation point. The factors consist of the second resonant frequencies in X-axis and Z-axis directions were designated as the major factors, and the residual vibration signals contributed to improving the evaluation effects of cabbage quality, as well as the morphological index, mass, density, and logarithmic damping ratio in the Z-axis. The proposed approach provides a method for non-destructive detection of Chinese cabbage quality, and it may be applied to online inspection, grading, and quality judgment during the post-harvest process.

Comparison of methods for filling in gaps in monthly rainfall series in arid regions

AbstractFull and reliable rainfall data are required for hydrological modelling, integrated water resource management, and planning. However, these data suffer from record gaps and sparse rain gage distribution, which implies that the use of an imputation method is crucial. This study aims to compare the outputs from 10 imputation methods that were used to infill missing rainfall depth data (MRD) in an arid Mediterranean region. Different statistical tests were used to assess the outputs from the imputation methods. The results showed that for MRD ranges between 5 and 20% the stepwise multiple linear regression (MLRsw) method was valid and produced the best results with a root‐mean‐square error (RSME) and mean absolute error (MAE) of less than 7 and 2 mm, respectively. This was followed by the Monte Carlo Markov chain expectation–maximization‐based multiple imputation (MI‐MCMC) method, which had an RSME and MAE of 1.01 and 0.08 mm, respectively, at 20% MRD. On the other hand, the use of satellite data for imputation (LR_GPCC estimates) was appropriate for MRD ranging between 10 and 15%, while the statistical and spatial method was suitable for MRD of less than 5%.