Development of models predicting biodegradation rate rating with multiple linear regression and support vector machine algorithms

Integrated Transfer Learning and Multitask Learning Strategies to Construct Graph Neural Network Models for Predicting Bioaccumulation Parameters of Chemicals

Accurate prediction of parameters related to the environmental exposure of chemicals is crucial for the sound management of chemicals. However, the lack of large data sets for training models may result in poor prediction accuracy and robustness. Herein, integrated transfer learning (TL) and multitask learning (MTL) was proposed for constructing a graph neural network (GNN) model (abbreviated as TL-MTL-GNN model) using n-octanol/water partition coefficients as a source domain. The TL-MTL-GNN model was trained to predict three bioaccumulation parameters based on enlarged data sets that cover 2496 compounds with at least one bioaccumulation parameter. Results show that the TL-MTL-GNN model outperformed single-task GNN models with and without the TL, as well as conventional machine learning models trained with molecular descriptors or fingerprints. Applicability domains were characterized by a state-of-the-art structure-activity landscape-based (abbreviated as ADSAL) methodology. The TL-MTL-GNN model coupled with the optimal ADSAL was employed to predict bioaccumulation parameters for around 60,000 chemicals, with more than 13,000 compounds identified as bioaccumulative chemicals. The high predictive accuracy and robustness of the TL-MTL-GNN model demonstrate the feasibility of integrating the TL and MTL strategy in modeling small-sized data sets. The strategy holds significant potential for addressing small data challenges in modeling environmental chemicals.

Unveiling potent inhibitors for schistosomiasis through ligand-based drug design, molecular docking, molecular dynamics simulations and pharmacokinetics predictions

Schistosomiasis is a neglected tropical disease which imposes a considerable and enduring impact on affected regions, leading to persistent morbidity, hindering child development, diminishing productivity, and imposing economic burdens. Due to the emergence of drug resistance and limited management options, there is need to develop additional effective inhibitors for schistosomiasis. In view of this, quantitative structure-activity relationship studies, molecular docking, molecular dynamics simulations, drug-likeness and pharmacokinetics predictions were applied to 39 Schistosoma mansoni Thioredoxin Glutathione Reductase (SmTGR) inhibitors. The chosen QSAR model demonstrated robust statistical parameters, including an R2 of 0.798, R2adj of 0.767, Q2cv of 0.681, LOF of 0.930, R2test of 0.776, and cR2p of 0.746, confirming its reliability. The most active derivative (compound 40) was identified as a lead candidate for the development of new potential non-covalent inhibitors through ligand-based design. Subsequently, 12 novel compounds (40a-40l) were designed with enhanced anti-schistosomiasis activity and binding affinity. Molecular docking studies revealed strong and stable interactions, including hydrogen bonding, between the designed compounds and the target receptor. Molecular dynamics simulations over 100 nanoseconds and MM-PBSA free binding energy (ΔGbind) calculations validated the stability of the two best-designed molecules. Furthermore, drug-likeness and pharmacokinetics prediction analyses affirmed the potential of these designed compounds, suggesting their promise as innovative agents for the treatment of schistosomiasis.

An in-vitro three-dimensional surgical simulation technique to predict tibial tunnel length in transtibial posterior cruciate ligament reconstruction

BACKGROUND

During the transtibial posterior cruciate ligament (PCL) reconstruction, drilling depth excessively longer than the tibial tunnel length (TTL) is an important reason to cause popliteal neurovascular bundle injury when preparing the tibial tunnel. This study aims to develop an in-vitro three-dimensional surgical simulation technique to determine the TTL in anteromedial (AM) and anterolateral (AL) approaches.

METHODS

A total of 63 knees' 3-dimensional (3D) computed tomography models were included in this study. The SuperImage system was used to reconstruct the 3D knee model and locate the tibial PCL site. The established 3D knee model and the coordinates of the tibial PCL site were imported into Rhinoceros 3D modeling software to simulate AM and AL tibial tunnel approaches with different tibial tunnel angles (TTA). The TTL and the tibial tunnel height (TTH) were measured in this study.

RESULTS

In AM and AL tibial tunnel approaches, the TTL showed a strong correlation with the TTA (for AM: r = 0.758, p < 0.001; for AL: r = 0.727, p < 0.001). The best fit equation to calculate the TTL based on the TTA was Y = 1.04X + 14.96 for males in AM approach, Y = 0.93X + 17.76 for males in AL approach, Y = 0.92X + 14.4 for females in AM approach, and Y = 0.94X + 10.5 for females in AL approach.

CONCLUSION

Marking the TTL on the guide pin or reamer could help to avoid the drill bit over-penetrated into the popliteal space to damage the neurovascular structure.

Etemadi regression in chemometrics: Reliability-based procedures for modeling and forecasting

The creation of predictive models with a high degree of generalizability in chemical analysis and process optimization is of paramount importance. Nonetheless, formulating a prediction model based on collected data from chemical measurements that maximize quantitative generalizability remains a challenging task for chemometrics experts. To tackle this challenge, a range of forecasting models with varying characteristics, structures, and capabilities has been developed, utilizing either accuracy-based or reliability-based modeling strategies. While the majority of models follow the accuracy-based approach, a recently proposed reliability-based approach, known as the Etemadi approach, has shown impressive performance across various scientific fields. The Etemadi models were constructed through a reliability-based parameter estimation process in such a manner that maximizes the models' reliability. However, the foundation of modeling procedures for chemometrics purposes is built upon the assumption that high generalizability in inaccessible/test data is achieved through the accuracy-based training procedure in which errors in available historical/training data are minimized. After conducting a thorough review of the current literature, we have found that none of the forecasting models for chemometrics purposes incorporate reliability into their modeling procedures. Given the dynamic and highly sensitive nature of chemistry experiments and processes, implementing a reliable model that controls performance criteria variation is a promising strategy for achieving stable and robust forecasts. To address this research gap, this paper introduces several key innovations, which can be highlighted as follows: (1) Proposing a general design structure based on a new optimal reliability-based parameter estimation process. (2) Introducing a novel risk-based modeling strategy that minimizes the performance variation of models implemented under different conditions in chemical laboratory experiments, to generate a more generalizable model for diverse applications in chemometrics. (3) Specifying the degree of influence that each reliability and accuracy factor has in enhancing the generalizability and uncertainty modeling of chemometric models. Empirical evidence confirms the effectiveness and superior performance of reliability-based models compared to accuracy-based models in 78.95% of the cases across various fields, including Pharmacology, Biochemistry, Agrochemical, Geochemical, Biological, Pollutants, Physicochemical Properties, and Gases Experiment. Furthermore, the study's findings demonstrate that the reliability-based modeling approach outperforms the accuracy-based strategy in terms of MAE, MSE, ARV, and RMSE by an average of 4.697%, 5.646%, 5.646%, and 4.342%, respectively. It is also statistically proven that reliability has a more significant impact on improving the generalizability of chemometric models than accuracy. This emphasizes the importance of including reliability as a crucial factor in chemometrics modeling, a consideration that has been overlooked in traditional modeling processes. Consequently, reliability-based modeling approaches can be regarded as a viable alternative to conventional accuracy-based modeling methods for chemical modeling purposes.

Prioritization of chemicals in personal care products based on persistent, bioaccumulative and toxic (PBT) potential: An Indian perspective

Numerous organic ingredients present in Personal care products (PCPs) are being detected in sewage which has a high potential to impact the environment. These compounds are called as Emerging contaminants (ECs) or Contaminants of emerging concern. However, the information on the source and occurrence of ECs present in PCPs is very minimal. Specifically, information on the persistence (P), bioaccumulation (B) and toxicity (T) is very scarce. The determination of PBT properties is a complex task given the magnitude of chemicals, thus it is necessary to have a tool to quickly screen and prioritize the most important compounds. Estimation Program Interface (EPI) Suite™ is one such tool authorized by United States Environmental Protection Agency (US EPA) for screening purposes. In the present study, several organic compounds present in the PCPs viz. body lotion, sunscreens, moisturizers, hair dyes, and some hair care products were identified and their PBT property was estimated. The results from the study indicate that the order of increasing occurrence of PBT chemicals is sunscreens > moisturizers > body lotion > hair care products > hair dyes. The prioritized compounds were given rank 1 (maximum concern) to rank 4 (minimum concern). From the results, the compounds octocrylene in sunscreens and butylphenyl methylpropional in body lotions were prioritized as Rank 1.

Improving predictions and understanding of primary and ultimate biodegradation rates with machine learning models

This study aimed to develop machine learning based quantitative structure biodegradability relationship (QSBR) models for predicting primary and ultimate biodegradation rates of organic chemicals, which are essential parameters for environmental risk assessment. For this purpose, experimental primary and ultimate biodegradation rates of high consistency were compiled for 173 organic compounds. A significant number of descriptors were calculated with a collection of quantum/computational chemistry software and tools to achieve comprehensive representation and interpretability. Following a pre-screening process, multiple QSBR models were developed for both primary and ultimate endpoints using three algorithms: extreme gradient boosting (XGBoost), support vector machine (SVM), and multiple linear regression (MLR). Furthermore, a unified QSBR model was constructed using the knowledge transfer technique and XGBoost. Results demonstrated that all QSBR models developed in this study had good performance. Particularly, SVM models exhibited high level of goodness of fit (coefficient of determination on the training set of 0.973 for primary and 0.980 for ultimate), robustness (leave-one-out cross-validated coefficient of 0.953 for primary and 0.967 for ultimate), and external predictive ability (external explained variance of 0.947 for primary and 0.958 for ultimate). The knowledge transfer technique enhanced model performance by learning from properties of two biodegradation endpoints. Williams plots were used to visualize the application domains of the models. Through SHapley Additive exPlanations (SHAP) analysis, this study identified key features affecting biodegradation rates. Notably, MDEO-12, APC2D1_C_O, and other features contributed to primary biodegradation, while AATS0v, AATS2v, and others inhibited it. For ultimate biodegradation, features like No. of Rotatable Bonds, APC2D1_C_O, and minHBa were contributors, while C1SP3, Halogen Ratio, GGI4, and others hindered the process. Also, the study quantified the contributions of each feature in predictions for individual chemicals. This research provides valuable tools for predicting both primary and ultimate biodegradation rates while offering insights into the mechanisms.

Predictive in silico models for aquatic toxicity of cosmetic and personal care additive mixtures

As emerging environmental contaminants, cosmetic and personal care additives (CPCAs) may have less oversight than other consumer products. Their continuous release and pseudopersistence could cause long-term harm to the aquatic environment. Since CPCAs generally exist in the form of mixtures in the environment, prediction and analysis of their mixture toxicity are crucial for ecological risk assessment. In this study, the acute toxicity of five typical CPCA mixtures to Daphnia magna was tested. The combined toxicity of binary mixtures was examined with the traditional concentration addition (CA) and independent action (IA) model. Overall, the synergistic effect of the five CPCAs may be caused mainly by methylparaben. In addition, reliable approaches for quantitative structure-activity relationship (QSAR) model development were explored. Specifically, 18 QSAR models were developed by three dataset partitioning techniques (Kennard-Stone's algorithm division, Euclidean distance based division, and sorted activity based division), two descriptor filtering methods (genetic algorithm and stepwise multiple linear regression) and three regression methods (multiple linear regression, partial least squares and support vector machine). Sixteen equations were applied for the calculation of the mixture descriptors to screen the functional expression of the mixture descriptors with the largest contribution to the mixture toxicity. A new comprehensive parameter that integrates internal and external validation was proposed for QSAR models evaluation. The mixture toxicity is mainly related the 3D distribution of atomic masses and the spatial distribution of the molecule electronic properties. Rigorously validated and externally predictive QSAR models were developed for predicting the toxicity of binary CPCAs mixtures with any ratio, in the applicability domain. The best possible work frame for construction and validation of QSAR models to provide reliable predictions on the mixture toxicity was proposed.

Experiments and machine learning-based modeling for haloacetic acids rejection by nanofiltration: Influence of solute properties and operating conditions

Because of potential risks to public health, the presence of haloacetic acids (HAAs) in drinking water is a major concern. Nanofiltration (NF) has shown potential for HAAs rejection, and several factors, namely, membrane properties, solute properties, and operating conditions, have been revealed key roles. However, knowledge of NF separation mechanism by quantifying these factors is limited. This study investigated and modeled NF performance on HAAs rejection. NF performance was experimentally investigated under various transmembrane pressure (TMP), cross-flow velocity (CV), temperature, pH, ionic strength (IS), and HAAs initial feed concentration (C_in). We used machine learning (ML) to understand the mechanism from the perspective of HAAs properties and operating conditions. Multiple linear regression (MLR), support vector machine (SVM), multsilayer perceptron (MLP), extreme gradient boosting (XGBoost), and random forest (RF) models were used. The MLP, XGBoost and RF models achieved significant performance with high R² (0.970, 0.973, and 0.980) and low RMSE (4.71, 4.41, and 3.84). These three models were analyzed using the Shapley Additive explanation (SHAP) to quantify relative contributions of HAAs properties and operating conditions. XGBoost-SHAP produced the most logical results and was the best-performing model for selecting optimal input variables combinations. The results showed that Stokes radius (r_s), logarithmic octanol-water partitioning coefficient (logK_ow), molecular weight (MW), pH, TMP, and temperature are key variables for interpreting NF process. The effects of HAAs properties were ranked as r_s > logK_ow > MW, suggesting significance of size exclusion and hydrophobic interaction. The impact of the operational conditions followed the order pH > TMP > temperature, illustrating that pH was the major influencing operating condition. This study demonstrated significant capacity of ML, which reduced amount of experimental work. In addition, the main operating conditions can be evaluated in terms of their contributions, making ML an efficient tool for risk management and process optimization.

Comparative Analysis of the Implementation of Support Vector Machines and Long Short-Term Memory Artificial Neural Networks in Municipal Solid Waste Management Models in Megacities

The development of methodologies to support decision-making in municipal solid waste (MSW) management processes is of great interest for municipal administrations. Artificial intelligence (AI) techniques provide multiple tools for designing algorithms to objectively analyze data while creating highly precise models. Support vector machines and neuronal networks are formed by AI applications offering optimization solutions at different managing stages. In this paper, an implementation and comparison of the results obtained by two AI methods on a solid waste management problem is shown. Support vector machine (SVM) and long short-term memory (LSTM) network techniques have been used. The implementation of LSTM took into account different configurations, temporal filtering and annual calculations of solid waste collection periods. Results show that the SVM method properly fits selected data and yields consistent regression curves, even with very limited training data, leading to more accurate results than those obtained by the LSTM method.

Prediction of organic contaminant rejection by nanofiltration and reverse osmosis membranes using interpretable machine learning models

Efficiency improvement in contaminant removal by nanofiltration (NF) and reverse osmosis (RO) membranes is a multidimensional process involving membrane material selection and experimental condition optimization. It is unrealistic to explore the contributions of diverse influencing factors to the removal rate by trial-and-error experimentation. However, the advanced machine learning (ML) method is a powerful tool to simulate this complex decision-making process. Here, 4 traditional learning algorithms (MLR, SVM, ANN, kNN) and 4 ensemble learning algorithms (RF, GBDT, XGBoost, LightGBM) were applied to predict the removal efficiency of contaminants. Results reported here demonstrate that ensemble models showed significantly better predictive performance than traditional models. More importantly, this study achieved a compelling tradeoff between accuracy and interpretability for ensemble models with an effective model interpretation approach, which revealed the mutual interaction mechanism between the membrane material, contaminants and experimental conditions in membrane separation. Additionally, feature selection was for the first time achieved based on the aforementioned model interpretation method to determine the most important variable influencing the contaminant removal rate. Ultimately, the four ensemble models retrained by the selected variables achieved distinguished prediction performance (R²_adj = 92.4 %-99.5 %). MWCO (membrane molecular weight cut-off), McGowan volume of solute (V) and molecular weight (MW) of the compound were demonstrated to be the most important influencing factors in contaminant removal by the NF and RO processes. Overall, the proposed methods in this study can facilitate versatile complex decision-making processes in the environmental field, particularly in contaminant removal by advanced physicochemical separation processes.

Pathological and ATR-FTIR spectral changes of delayed splenic rupture and medical significance

Traumatic delayed splenic rupture often follows by a "latent period" without typical symptoms after injury. During this period, though there are no obvious symptoms, the injury is still present and changing. In this study, we constructed an SD rat model of delayed splenic rupture; evaluated the model by HE staining, Perl's staining, Masson trichrome staining and immunohistochemical staining; observed the pathological changes of spleen tissue in delayed splenic rupture at different times after splenic injury; we found that pathological change of injured tissues were different from non-injured, and has phases-change patterns, it can be roughly divided into three phases: 2-7 d, 10-14 d, and 18-28.We then investigated the relationship between the pathological changes and FTIR spectroscopy by chemometric methods. The main distinction of injured and non-injured tissue was the protein secondary structure of amide I, and the main distinctions of different phases of delayed splenic rupture were protein secondary structures and content of amide I and amide II.A classification model developed by SVM-DA was used to infer three phases (2-7 days, 10-12 days and 14-28 days). According to the most probable class, the accuracy of external validation is 96.7%. The results indicate that FTIR spectroscopy combined with various types of pathological staining has a potential for forensic identification and can provide theoretical support and diagnostic reference on clinical persistent injury.

Classification and Regression Machine Learning Models for Predicting Aerobic Ready and Inherent Biodegradation of Organic Chemicals in Water

Machine learning (ML) is viewed as a promising tool for the prediction of aerobic biodegradation, one of the most important elimination pathways of organic chemicals from the environment. However, available models only have small datasets (<3200 records), make binary classification predictions, evaluate ready biodegradability, and do not incorporate experimental conditions (e.g., system setup and reaction time). This study addressed all these limitations by first compiling a large database of 12,750 records, considering both ready and inherent biodegradation under different conditions, and then developing regression and classification models using different chemical representations and ML algorithms. The best regression model (R² = 0.54 and root mean square error of 0.25) and classification model (the prediction accuracy from 85.1%) achieved very good performance. The model interpretation indicated that the models correctly captured the effects of chemical substructures, following the order of C═O > O═C-O > OH > CH₃ > halogen > branching > N > 6-member ring. The consideration of chemical speciation based on pK_a and α notations did not affect the regression model performance but significantly improved the classification model performance (the accuracy increased to 87.6%). The models also showed large applicability domains and provided reasonable predictions for more than 98% of over 850,000 environmentally relevant chemicals in the Distributed Structure-Searchable Toxicity database. These robust, trustable models were finally made widely accessible through two free online predictors with graphical user interface.

Estimating the Physical Properties of Nanofluids Using a Connectionist Intelligent Model Known as Gaussian Process Regression Approach

This work aims to develop a robust machine learning model for the prediction of the relative viscosity of nanoparticles (NPs) including Al2O3, TiO2, SiO2, CuO, SiC, and Ag based on the most important input parameters affecting them covering the size, concentration, thickness of the interfacial layer, and intensive properties of NPs. In order to develop a comprehensive artificial intelligence model in this study, sixty-nine data samples were collected. To this end, the Gaussian process regression approach with four basic function kernels (Matern, squared exponential, exponential, and rational quadratic) was exploited. It was found that Matern outperformed other models with R2 = 0.987, MARE (%) = 6.048, RMSE = 0.0577, and STD = 0.0574. This precise yet simple model can be a good alternative to the complex thermodynamic, mathematical-analytical models of the past.

A novel reliability-based regression model to analyze and forecast the severity of COVID-19 patients

Background

Coronavirus outbreak (SARS-CoV-2) has become a serious threat to human society all around the world. Due to the rapid rate of disease outbreaks and the severe shortages of medical resources, predicting COVID-19 disease severity continues to be a challenge for healthcare systems. Accurate prediction of severe patients plays a vital role in determining treatment priorities, effective management of medical facilities, and reducing the number of deaths. Various methods have been used in the literature to predict the severity prognosis of COVID-19 patients. Despite the different appearance of the methods, they all aim to achieve generalizable results by increasing the accuracy and reducing the errors of predictions. In other words, accuracy is considered the only effective factor in the generalizability of models. In addition to accuracy, reliability and consistency of results are other critical factors that must be considered to yield generalizable medical predictions. Since the role of reliability in medical decisions is significant, upgrading reliable medical data-driven models requires more attention.

Methods

This paper presents a new modeling technique to specify and maximize the reliability of results in predicting the severity prognosis of COVID-19 patients. We use the well-known classic regression as the basic model to implement our proposed procedure on it. To assess the performance of the proposed model, it has been applied to predict the severity prognosis of COVID-19 by using a dataset including clinical information of 46 COVID-19 patients. The dataset consists of two types of patients’ outcomes including mild (discharge) and severe (ICU or death). To measure the efficiency of the proposed model, we compare the accuracy of the proposed model to the classic regression model.

Results

The proposed reliability-based regression model, by achieving 98.6% sensitivity, 88.2% specificity, and 93.10% accuracy, has better performance than classic accuracy-based regression model with 95.7% sensitivity, 85.5% specificity, and 90.3% accuracy. Also, graphical analysis of ROC curve showed AUC 0.93 (95% CI 0.88–0.98) and AUC 0.90 (95% CI 0.85–0.96) for classic regression models, respectively.

Conclusions

Maximizing reliability in the medical forecasting models can lead to more generalizable and accurate results. The competitive results indicate that the proposed reliability-based regression model has higher performance in predicting the deterioration of COVID-19 patients compared to the classic accuracy-based regression model. The proposed framework can be used as a suitable alternative for the traditional regression method to improve the decision-making and triage processes of COVID-19 patients.

Prediction models with multiple machine learning algorithms for POPs: The calculation of PDMS-air partition coefficient from molecular descriptor

Polydimethylsiloxane-air partition coefficient (K_PDMS-air) is a key parameter for passive sampling to measure POPs concentrations. In this study, 13 QSPR models were developed to predict K_PDMS-air, with two descriptor selection methods (MLR and RF) and seven algorithms (MLR, LASSO, ANN, SVM, kNN, RF and GBDT). All models were based on a data set of 244 POPs from 13 different categories. The diverse model evaluation parameters calculated from training and test set were used for internal and external verification. Notably, the R_adj², Q_BOOT² and Q_ext² are 0.995, 0.980 and 0.951 respectively for GBDT model, showing remarkable superiority in fitting, robustness and predictability compared with other models. The discovery that molecular size, branches and types of the bonds were the main internal factors affecting the partition process was revealed by mechanism explanation. Different from the existing QSPR models based on single category compounds, the models developed herein considered multiple classes compounds, so that its application domain was more comprehensive. Therefore, the obtained models can fill the data gap of missing experimental K_PDMS-air values for compounds in the application range, and help researchers better understand the distribution behavior of POPs from the perspective of molecular structure.

Energy information flow-based ecological risk transmission among communities within the heavy metals contaminated soil system

An energy information flow-based ecological risk assessment framework (EIF-ERA) is developed for identifying ecological risk transmission rules among communities (i.e., vegetation E1, herbivorous animals E2, soil microorganisms E3, and carnivorous animals E4) within the heavy metals contaminated soil system. This framework is integrated with numerous techniques of carcinogenic risk evaluation, ecological risk assessment (ERA), and Monte Carlo simulation. Stepwise quadratic response surface analysis (SQRSA) is employed for reflecting the relation between contaminants' concentration and comprehensive risk. Two scenarios with respect to the environmental quality standards (scenarios 1) and carcinogenic risk reversion (scenarios 2) are merged into the EIF-ERA. A real-world mining area in Xinglong County in Chengde is selected to verify the developed framework's effectiveness. Results reveal that E3 is considered as the most sensitive community when contaminant interference occurs, and its 62.3% and 37.7% of comprehensive risk are contributed by initial and direct risks, respectively. Other communities can receive direct risk through control allocation (CA). Monte Carlo anlysis shows that there are 7.68% and 20.25% increase in the initial risk of Cd and Pb when their quantile statistics increase from 70% to 90%. Determination of an appropriate screening value is vital for contaminated mining soil remediation due to its inefficiency of remediation funds, especially when considering the trict standards of contaminants' concentration within scenarios 1. The surrogates obtained from the SQRSA display the relation of contaminant concentration and comprehensive risks with the adjusted R² greater than 0.77. These findings can be in support of system design, risk assessment, and site remediation.

Development of novel experimental and modelled low density polyethylene (LDPE)-water partition coefficients for a range of hydrophobic organic compounds

Knowledge about partitioning constants of hydrophobic organic compounds (HOCs) between the polymer and aqueous phases is critical for assessing chemical environmental fate and transport. The conventional experimental method is characterized by large discrepancies in the measured values due to the limited water solubility of HOCs and other associated issues. In the current work, a novel three-phase partitioning system was evaluated to determine accurate low-density polyethylene (LDPE)-water partition coefficients (K_PE-w). By adding sufficient surfactant (Brij 30) to form the micellar pseudo-phase within the polymer/water system, the K_PE-w values were obtained from a combination of two experimentally measured values, that is, the micelle-water partition coefficient (K_mic-w) and the LDPE-micelle partition coefficient (K_PE-mic). The method presented here is capable of shortening the equilibration time to half a month, and avoiding defects of the traditional method with respect to directly measured aqueous phase concentrations. Herein, the K_PE-w values were determined for HOCs with little errors. Meanwhile, based on the 120 experimental K_PE-w data, several in silico models were also developed as valid extrapolation tools to estimate missing or uncertain values. Analysis of the underlying solubility interactions in the nonionic surfactant micelles were investigated, providing additional support for the reliability of the proposed method.

In silico prediction of polyethylene-aqueous and air partition coefficients of organic contaminants using linear and nonlinear approaches

Low-density polyethylene (LDPE) passive sampling is very attractive for use in determining chemicals concentrations. Crucial to the measurement is the coefficient (K_PE) describing partitioning between LDPE and environmental matrices. 255, 117 and 190 compounds were collected for the development of datasets in three different matrices, i.e., water, air and seawater, respectively. Further, 3 pp-LFER models and 9 QSPR models based on classical multiple linear regression (MLR) coupled with prevalent nonlinear algorithms (artificial neural network, ANN and support vector machine, SVM) were performed to predict LDPE-water (K_PE-W), LDPE-air (K_PE-A) and LDPE-seawater (K_PE-SW) partition coefficients. These developed models have satisfying predictability (R²_adj: 0.805-0.966, 0.963-0.991 and 0.817-0.941; RMSE_tra: 0.233-0.565, 0.200-0.406 and 0.260-0.459) and robustness (Q²_ext: 0.840-0.943, 0.968-0.984 and 0.797-0.842; RMSE_ext: 0.308-0.514, 0.299-0.426 and 0.407-0.462) in three datasets (water, air and seawater), respectively. In particular, the reasonable mechanism interpretations revealed that the molecular size, hydrophobicity, polarizability, ionization potential, and molecular stability were the most relevant properties, for governing chemicals partitioning between LDPE and environmental matrices. The application domains (ADs) assessed here exhibited the satisfactory applicability. As such, the derived models can act as intelligent tools to predict unknown K_PE values and fill the experimental gaps, which was further beneficial for the construction of enormous and reliable database to facilitate a distinct understanding of the distribution for organic contaminants in total environment.

Exploring QSPR models for predicting PUF-air partition coefficients of organic compounds with linear and nonlinear approaches

Partition coefficients are important parameters for measuring the concentration of chemicals by passive sampling devices. Considering the wide application of the polyurethane foam (PUF) in passive air sampling, an attempt for developing several quantitative structure-property relationship (QSPR) models was made in this work, to predict PUF-air partition coefficients (K_PUF-air) using linear (multiple linear regression, MLR) and non-linear (artificial neural network, ANN and support vector machine, SVM) methods by machine learning. All of the developed models were performed on a dataset of 170 compounds comprising 9 distinct classes. A series of statistical parameters and validation results showed that models had good prediction ability, robustness and goodness-of-fit. Furthermore, the underlying mechanisms of molecular descriptors emphasized that ionization potential, molecular bond, hydrophilicity, size of molecule and valence electron number had dominating influence on the adsorption process of chemicals. Overall, the obtained models were all established on the extensive applicability domains, and thus can be used as effective tools to predict the K_PUF-air of new organic compounds or those have not been synthesized yet which, in turn, could help researchers better understand the mechanistic basis of adsorption behavior of PUF.

Classification of Biodegradable Substances Using Balanced Random Trees and Boosted C5.0 Decision Trees

Substances that do not degrade over time have proven to be harmful to the environment and are dangerous to living organisms. Being able to predict the biodegradability of substances without costly experiments is useful. Recently, the quantitative structure-activity relationship (QSAR) models have proposed effective solutions to this problem. However, the molecular descriptor datasets usually suffer from the problems of unbalanced class distribution, which adversely affects the efficiency and generalization of the derived models. Accordingly, this study aims at validating the performances of balanced random trees (RTs) and boosted C5.0 decision trees (DTs) to construct QSAR models to classify the ready biodegradation of substances and their abilities to deal with unbalanced data. The balanced RTs model algorithm builds individual trees using balanced bootstrap samples, while the boosted C5.0 DT is modeled using cost-sensitive learning. We employed the two-dimensional molecular descriptor dataset, which is publicly available through the University of California, Irvine (UCI) machine learning repository. The molecular descriptors were ranked according to their contributions to the balanced RTs classification process. The performance of the proposed models was compared with previously reported results. Based on the statistical measures, the experimental results showed that the proposed models outperform the classification results of the support vector machine (SVM), K-nearest neighbors (KNN), and discrimination analysis (DA). Classification measures were analyzed in terms of accuracy, sensitivity, specificity, precision, false positive rate, false negative rate, F1 score, receiver operating characteristic (ROC) curve, and area under the ROC curve (AUROC).