Quantification of the antagonistic and synergistic effects of Pb2+, Cu2+, and Zn2+ bioaccumulation by living Bacillus subtilis biomass using XGBoost and SHAP

Non-targeted metabolomics and explainable artificial intelligence: Effects of processing and color on coniferyl aldehyde levels in Eucommiae cortex

Eucommia ulmoides, a plant native to China, is valued for its medicinal properties and has applications in food, health products, and traditional Chinese medicine. Processed Eucommiae Cortex (EC) has historically been a highly valued medicine. Ancient doctors had ample experience processing EC, especially with ginger juice, as documented in traditional Chinese medical texts. The combination of EC and ginger juice helps release and transform the active ingredients, strengthening the medicine's effectiveness and improving its taste and shelf life. However, the lack of quality control standards for Ginger-Eucommiae Cortex (G-EC), processed from EC and ginger, presents challenges for its industrial and clinical use. This study optimized G-EC processing using the CRITIC and Box-Behnken methods. Metabolomics showed 517 chemical changes between raw and processed G-EC, particularly an increase in coniferyl aldehyde (CFA). Explainable artificial intelligence techniques revealed the feasibility of using color to CFA content, providing insights into quality indicators.

Assessment and simulation of eco-environmental quality changes in rapid rural urbanization: Xiong'an New Area, China

Xiong'an New Area was established as a state-level new area in 2017 and serves as a typical representative area for studying the ecological evolution of rural areas under rapid urbanization in China. Remote sensing-based ecological index (RSEI) is a regional eco-environmental quality (EEQ) assessment index. Many studies have employed RSEI to achieve rapid, objective, and effective quantitative assessment of the spatio-temporal changes of regional EEQ. However, research that combines RSEI with machine learning algorithms to conduct multi-scenario simulation of EEQ is still relatively scarce. Therefore, this study assessed and simulated EEQ changes in Xiong'an and revealed that: (1) The large-scale construction has led to an overall decline in EEQ, with the RSEI decreasing from 0.648 in 2014 to 0.599 in 2021. (2) Through the multi-scenario simulation, the non-unidirectional evolution of RSEI during the process of urban-rural construction has been revealed, specifically characterized by a significant decline followed by a slight recovery. (3) The marginal effects of urban-rural construction features for simulated RSEI demonstrate an inverted "U-shaped" curve in the relationship between urbanization and EEQ. This indicates that urbanization and EEQ may not be absolute zero-sum. These findings can provide scientific insights for maintaining and improving the regional EEQ in urban-rural construction.

Dual-recognition "turn-off-on" fluorescent Biosensor triphenylamine-based continuous detection of copper ion and glyphosate applicated in environment and living system

Heavy metal Cu2+ emitted in industry and residues of glyphosate pesticides are pervasive in ecosystems, accumulated in water bodies and organisms' overtime, constituting hazard to human and ecological balance. The development of rapid, highly selective, reversibility and sensitive biosensor in vivo detection for Cu2+ and glyphosate was imminent. A novel dual-recognition fluorescence biosensor MPH was successfully synthesized based on triphenylamine, which demonstrated remarkable ratiometric fluorescence quenching toward Cu2+, while MPH-Cu2+ (1:1) ensemble exhibited ratiometric fluorescence restoration for glyphosate, both with observable color changes in daylight and UV lamp. The biosensor exhibited rapid, outstanding selectivity, anti-interference, and multiple cycles reversibility through "turn-off-on" fluorescence towards Cu2+ and glyphosate, respectively. Surprisingly, the clearly binding mechanisms of MPH to Cu2+ and MPH-Cu2+ ensemble to glyphosate were determined, respectively, based on the Job's plot, FT-IR, ESI-HRMS, 1H NMR titration and theoretical calculations of dynamics and thermodynamics. In addition, biosensor MPH demonstrated successful detection of Cu2+ and glyphosate across diverse environmental samples including tap water, extraction solutions of traditional Chinese medicine honeysuckle and soil samples. In the meantime, fluorescence imaging of Cu2+ and glyphosate at both micro and macro scales in various living organisms, such as rice roots, MCF-7 cells, zebrafish, and mice, were successfully achieved. Overall, this work was expected to become a promising and versatile fluorescence biosensor for rapid and reversible detection of Cu2+ and glyphosate both in vitro and vivo.

Prediction of micropollutant degradation kinetic constant by ultrasonic using machine learning

A prediction model based on XGBoost is proposed for ultrasonic degradation of micropollutants' kinetic constants. After parameter optimization through iteration, the model achieves Evaluation metrics with R2 and SMAPE reaching 0.99 and 2.06%, respectively. The impact of design parameters on predicting kinetic constants for ultrasound degradation of trace pollutants was assessed using Shapley additive explanations (SHAP). Results indicate that power density and frequency significantly impact the predictive performance. The database was sorted based on power density and frequency values. Subsequently, 800 raw data were split into small databases of 200 each. After confirming that reducing the database size doesn't affect prediction accuracy, ultrasound degradation experiments were conducted for five pollutants, yielding experimental data. A small database with experimental conditions within the numerical range was selected. Data meeting both feature conditions were filtered, resulting in an optimized 60-data group. After incorporating experimental data, a model was trained for prediction. Degradation kinetic constants for experiments (kE) were compared with predicted constants (for 800 data-based model: kP-800 and for 60 data-based model: kP-60). Results showed ibuprofen, bisphenol A, carbamazepine, and 17β-Estradiol performed better on the 60-data group (kP-60/kE: 1.00, 0.99, 1.00, 1.00), while caffeine suited the model trained on the 800-data group (kP-800/kE: 1.02).

Innovative optimization for enhancing Pb2+ biosorption from aqueous solutions using Bacillus subtilis

Introduction

Toxic heavy metal pollution has been considered a major ecosystem pollution source. Unceasing or rare performance of Pb2+ to the surrounding environment causes damage to the kidney, nervous, and liver systems. Microbial remediation has acquired prominence in recent decades due to its high efficiency, environment-friendliness, and cost-effectiveness.

Methods

The lead biosorption by Bacillus subtilis was optimized by two successive paradigms, namely, a definitive screening design (DSD) and an artificial neural network (ANN), to maximize the sorption process.

Results

Five physicochemical variables showed a significant influence (p < 0.05) on the Pb2+ biosorption with optimal levels of pH 6.1, temperature 30°C, glucose 1.5%, yeast extract 1.7%, and MgSO4.7H2O 0.2, resulting in a 96.12% removal rate. The Pb2+ biosorption mechanism using B. subtilis biomass was investigated by performing several analyses before and after Pb2+ biosorption. The maximum Pb2+ biosorption capacity of B. subtilis was 61.8 mg/g at a 0.3 g biosorbent dose, pH 6.0, temperature 30°C, and contact time 60 min. Langmuir's isotherm and pseudo-second-order model with R2 of 0.991 and 0.999 were suitable for the biosorption data, predicting a monolayer adsorption and chemisorption mechanism, respectively.

Discussion

The outcome of the present research seems to be a first attempt to apply intelligence paradigms in the optimization of low-cost Pb2+ biosorption using B. subtilis biomass, justifying their promising application for enhancing the removal efficiency of heavy metal ions using biosorbents from contaminated aqueous systems.

Causal inference and mechanism for unraveling the removal of four pesticides from lettuce (Lactuca sativa L.) via ultrasonic processing and various immersion solutions

This study explores the reduction of carbamates (CAs) and pyrethroids (PYs) - commonly used pesticides - in lettuce using various immersion solutions and ultrasonic processing. It also examines the role of machine learning and molecular docking in understanding the mechanisms of pesticide reduction. The results revealed that the highest reduction of both CAs and PYs exceeded 80 % on lettuce leaves. In most samples, the reduction increased with the power of ultrasonic processing and processing time. The results of machine learning models (XGBoost and SHAP) showed that during the immersion cleaning of CAs and PYs, as well as during both immersion cleaning and ultrasonic processing of CAs + PYs, the reduction was most influenced by the initial pesticide levels and immersion time. Gas Chromatography-Mass Spectrometry (GC-MS) analysis of lettuce's wax layer identified 24 compounds, including fatty alcohols, fatty acids, fatty acid esters, and triterpenoids. Despite the absence of active sites, the lipophilic nature of long-chain aliphatic compounds aids in pesticide binding, while triterpenoids form strong hydrogen bonds with pesticides, indicating a robust adsorption on the lettuce surface. This study aims to offer insights into the efficient removal of chemical pesticide residues from fruits and vegetables, addressing critical concerns for food safety and human health.

Adsorption mechanism of Cd2+ on microbial inoculant and its potential for remediation Cd-polluted farmland soils

The adsorption characteristics and mechanism of Cd2+ on microbial inoculant (MI) mainly composed of Bacillus subtilis, Bacillus thuringiensis and Bacillus amyloliquefaciens, and its potential for remediation Cd polluted soils through batch adsorption and soil incubation experiments. It was found that the Freundlich isotherm model and the pseudo-second-order kinetics were more in line with the adsorption processes of Cd2+. The maximum adsorption capacity predicted by Langmuir isotherm model suggested that of MI was 57.38 mg g-1. Scanning electron microscopy and energy dispersive spectroscopy (SEM-EDS) images exhibited the surface structure of MI was damaged to varying degrees after adsorption, and Cd element was distributed on the surface of MI through ion exchange. X-ray diffraction (XRD) results showed that CdCO3 was formed on the surface of MI. Moreover, the functional groups (-OH, C-H, and -NH) involved in the adsorption of Cd2+ through fourier transform infrared spectroscopy (FTIR). After applying MI to Cd-contaminated soil, it was found that soil pH, conductivity (EC) and soil organic matter (SOM) increased by 0.84 %-2.43 %, 31.6 %-241.48 %, and 8.11 %-24.1 %, respectively, when compared with the control treatments. The content of DTPA-Cd in the soils was significantly (P < 0.05) reduced by 15.48 %-29.68 % in contrast with CK, and the Cd speciation was transformed into a more stable residual fraction. The activities of urease, phosphatase and sucrose were increased by 3.5 %-45.18 %, 57.00 %-134.18 % and 52.51 %-70.52 %, respectively, compared with CK. Therefore, MI could be used as an ecofriendly and sustainable material for bioremediation of Cd-contaminated soils.

Utilizing machine learning for reactive material selection and width design in permeable reactive barrier (PRB)

Permeable reactive barrier (PRB) is an important groundwater treatment technology. However, selecting the optimal reactive material and estimating the width remain critical and challenging problems in PRB design. Machine learning (ML) has advantages in predicting evolution and tracing contaminants in temporal and spatial distribution. In this study, ML was developed to design PRB, and its feasibility was validated through experiments and a case study. ML algorithm showed a good prediction about the Freundlich equilibrium parameter (R2 0.94 for KF, R2 0.96 for n). After SHapley Additive exPlanation (SHAP) analysis, redefining the range of the significant impact factors (initial concentration and pH) can further improve the prediction accuracy (R2 0.99 for KF, R2 0.99 for n). To mitigate model bias and ensure comprehensiveness, evaluation index with expert opinions was used to determine the optimal material from candidate materials. Meanwhile, the ML algorithm was also applied to predict the width of the mass transport zone in the adsorption column. This procedure showed excellent accuracy with R2 and root-mean-square-error (RMSE) of 0.98 and 1.2, respectively. Compared with the traditional width design methodology, ML can enhance design efficiency and save experiment time. The novel approach is based on traditional design principles, and the limitations and challenges are highlighted. After further expanding the data set and optimizing the algorithm, the accuracy of ML can make up for the existing limitations and obtain wider applications.

Prediction of microplastic abundance in surface water of the ocean and influencing factors based on ensemble learning

Microplastics are regarded as emergent contaminants posing a serious threat to the marine ecosystem. It is time-consuming and labor-intensive to determine the number of microplastics in different seas using traditional sampling and detection methods. Machine learning can provide a promising tool for prediction, but there is a lack of research on this. To screen high-performance models for the prediction of microplastic abundance in the marine surface water and explore the influencing factors, three ensemble learning models, random forest (RF), gradient boosted decision tree (GBDT), and extreme gradient boosting (XGBoost), were developed and compared. A total of 1169 samples were collected, and multi-classification prediction models were constructed with 16 features of the data as inputs and six classes of microplastic abundance intervals as outputs. Our results show that the XGBoost model has the best performance of prediction, with a total accuracy rate of 0.719 and an ROC AUC (Receiver Operating Characteristic curve, Area Under Curve) value of 0.914. Seawater phosphate (PHOS) and seawater temperature (TEMP) have negative effects on the abundance of microplastics in surface seawater, while the distance between the sampling point and the coast (DIS), wind stress (WS), human development index (HDI), and sampling latitude (LAT) have positive effects. This work not only predicts the abundance of microplastics in different seas but also offers a framework for the use of machine learning in the study of marine microplastics.

Zero-valent iron based materials selection for permeable reactive barrier using machine learning

The zero-valent iron (ZVI) based reactive materials are potential remediation reagents in permeable reactive barriers (PRB). Considering that reactive materials is the essential to determining the long-term stability of PRB and the emergence of a large number of new iron-based materials. Here, we present a new approach using machine learning to screen PRB reactive materials, which proposes to improve the efficiency and practicality of selection of ZVI-based materials. To compensate for the insufficient amount of existing machine learning source data and the real-world implementation, machine learning combines evaluation index (EI) and reactive material experimental evaluations. XGboost model is applied to estimate the kinetic data and SHAP is used to improve the accuracy of model. Batch and column tests were conducted to investigate the geochemical characteristics of groundwater. The study find that specific surface area is a fundamental factor correlated with the kinetic constants of ZVI-based materials, according to SHAP analysis. Reclassifying the data with specific surface area significantly improved prediction accuracy (reducing RMSE from 1.84 to 0.6). Experimental evaluation results showed that ZVI had 3.2 times higher anaerobic corrosion reaction kinetic constants and 3.8 times lower selectivity than AC-ZVI. Mechanistic studies revealed the transformation pathways and endpoint products of iron compounds. Overall, this study is a successful initial attempt to use machine learning for selecting reactive materials.

Use of Chrysosporium/carbon nanotubes for preconcentration of ultra-trace cadmium levels from various samples after extensive studies on its adsorption properties

Carbon nanotubes were used to immobilize Chrysosporium fungus for building an adequate adsorbent to be used as an desirable sorbent for preconcentration and measurement of cadmium ultra-trace levels in various samples. After characterization, the potential of Chrysosporium/carbon nanotubes for the sorption of Cd(II) ions was scrutinized by the aid of central composite design, and comprehensive studies of sorption equilibrium, kinetics and thermodynamic aspects were accomplished. Then, the composite was utilized for preconcentration of ultra-trace cadmium levels, by a mini-column packed with Chrysosporium/carbon nanotubes, before its determination with ICP-OES. The outcomes vouchsafed that (i) Chrysosporium/carbon nanotube has a high tendency for selective and rapid sorption of cadmium ion, at pH 6.1, and (ii) kinetic, equilibrium, and thermodynamic studies showed a high affinity of the Chrysosporium/carbon nanotubes for cadmium ion. Also, the outcomes displayed that cadmium can quantitatively be sorbed at a flow speed lesser than 7.0 mL/min and a 1.0 M HCl solution (3.0 mL) was sufficient to desorbe the analyte. Eventually, preconcentration and measurement of Cd(II) in different foods and waters were successfully accomplished with good accuracy, high precision (RSDs ≤5.65%), and low limit of detection (0.015 μg/L).