Prediction of chemical biodegradability using computational methods

Zhan, Zhixiong; Li, Linlang; Tian, Sheng; Zhen, Xuechu; Li, Youyong

doi:10.1080/08927022.2017.1328556

Cited by 12 publications

(17 citation statements)

References 60 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…This demonstrates their ability to efficiently handle biodegradation data. Compared to the findings reported in the preceding research in [ 11 , 15 , 17 , 18 ], the results of the proposed models are better for both training and test subsets. The results highlight the ability of these two tree-based ensembles to accommodate different feature types and to model such QSAR nonlinear relationships.…”

Section: Resultscontrasting

confidence: 72%

“…In [ 15 , 17 , 18 ], the authors employed the UCI biodegradation dataset donated by Mansouri et al in [ 11 ]. In [ 15 ], the authors reduced the 41 molecular descriptors using principal components analysis (PCA) into four components.…”

Section: Literature Reviewmentioning

confidence: 99%

“…ANN achieved an accuracy of 0.77, sensitivity of 0.61, and specificity of 0.85. In [ 17 ], the author applied the Naïve Bayesian classification (NBC) approach and achieved AUCs 0.890, 0.921, and 0.901 for training, test, and external validation subsets. In [ 18 ], the authors applied PLS_DA with uncertainty estimation.…”

Section: Literature Reviewmentioning

confidence: 99%

See 2 more Smart Citations

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

Elsayad

Nassef

Al‐Dhaifallah

et al. 2020

IJERPH

View full text Add to dashboard Cite

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).

show abstract

Section: Resultscontrasting

confidence: 72%

Section: Literature Reviewmentioning

confidence: 99%

See 1 more Smart Citation

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

Elsayad

Nassef

Al‐Dhaifallah

et al. 2020

IJERPH

View full text Add to dashboard Cite

show abstract

“…27,49 Note that these models were developed to predict the probability/time frame of biodegradation based on the group contributions of different substituents and are the basis of the US EPA's software EPI Suite for predicting biodegradability. 18 The linear regression of these averaged SHAP values against the six sets of coefficients had R 2 values of 0.73, 0.82, 0.41, 0.37, 0.56, and 0.52, respectively, for the substituents on aromatic chemicals (Figure 3) and 0.12, 0.05, 0.70, 0.65, 0.92, and 0.91, respectively, for the substituents on aliphatic chemicals (Figure S13). The detailed coefficient and the average SHAP values can be found in Tables S9 and S10.…”

Section: ■ Results and Discussionmentioning

confidence: 99%

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

Huang

Zhang

2022

Environ. Sci. Technol.

View full text Add to dashboard Cite

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 (R 2 = 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 > CH 3 > 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.

show abstract

“…2 and in Table I, while 2D chemical structures, IUPAC names and SMILES (simplified molecular-input line-entry system) that were used for molecules entry input in the ADMET Predictor TM software of entactogen molecules and o-quinone metabolites are displayed in Supplementary material in Table S-1a and Table S-1b, respectively. Statistical measurement and the prediction accuracy of ADMET Predictor's models used in this study are displayed in Table S-2 to Table S-5 in Supplementary Material. ADMET-related descriptors relevant to environmental toxicity were computed using the following aquatic toxicity models: 1) TOX_FHM, the fathead minnow acute toxicity model based on lethal effects on Pimephales promelas (Minnow_LC 50 ) which predicts concentration in mg L -1 of a given compound that will kill 50 % of a population of minnows after an exposure time of 96 hours (40); 2) TOX_ATTP model which is based on the inhibition of protozoa species Tetrahymena pyriformis and predicts concentration in mmol L -1 of a given compound needed to inhibit 50 % growth of T. pyroformis (Th_pyr_pIGC 50 ) after approximately 40 hours exposure (41); 3) TOX_DM model which is based on lethal concentration (mg L -1 ) that results in the death of 50 % of Daphnia magna (water fleas) (Daphnia LC 50 ) after 48 hours (42); 4) TOX_BCF model, the environmental toxicity based on bioconcentration factor (BCF) which is the ratio of the chemical concentration in biota to that in water at steady state, as a result of absorption via the respiratory surface, i.e., describes the accumulation of pollutants partitioning from the aqueous phase into an organic phase (typically fish), BCF = (Concentration in organism)/(Concentration in environment) (43) and 5) TOX_BIODEG model, the percent of biodegradation (% BD = 100 × BOD/ThOD) based on which a compound is considered readily biodegradable (RB) if the biological oxygen demand (BOD) is greater than or equal to 60 % of the theoretical oxygen demand (ThOD), otherwise that compound is considered non-readily biodegradable (44).…”

Section: Methodsmentioning

confidence: 99%

Evaluation of phenylethylamine type entactogens and their metabolites relevant to ecotoxicology – a QSAR study

Takač

Magina

Takać³

2019

Acta Pharmaceutica

View full text Add to dashboard Cite

The impact of the selected entactogens and their o-quinone metabolites on the environment was explored in QSAR studies by the use of predicted molecular descriptors, ADMET properties and environmental toxicity parameters, i.e., acute toxicity in Tetrahymena pyriformis (TOX_ATTP) expressed as Th_pyr_pIGC50/mmol L−1, acute toxicity in Pimephales promelas, the fathead minnow (TOX_FHM) expressed as Minnow LC50/mg L−1, the acute toxicity in Daphnia magna (TOX_DM) expressed as Daphnia LC50/mg L−1 and bioconcentration factor (BCF). The formation of corresponding o-quinones via benzo-dioxo-lone ring, O-demethylenation was predicted as the main metabolic pathway for all entactogens except for 1-(2,2-difluorobenzo[d][1,3]dioxol-5-yl)propan-2-amine (DiFMDA). The least favourable ADMET profile was revealed for N-(1-(benzo[d][1,3]dioxol-5-yl)propan-2-yl)-O-methylhydroxylamine (MDMEO). QSAR studies revealed significant linear correlations between MlogP of entactogens and MlogP of o-quinone metabolites (R = 0.99), and Th_pyr_pIGC50/mmol L−1 (R = 0.94), also their MlogPs with Minnow_LC50/mg L−1 (R = 0.80 and R = 0.78), BCF (R = 0.86 and R = 0.82) and percentage of o-quinones’ yields (R = 0.73 and R = 0.80). Entactogens were predicted as non-biodegradable molecules, whereas the majority of their o-quinones were biodegradable.

show abstract

Prediction of chemical biodegradability using computational methods

Cited by 12 publications

References 60 publications

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

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

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

Evaluation of phenylethylamine type entactogens and their metabolites relevant to ecotoxicology – a QSAR study

Contact Info

Product

Resources

About