Machine learning-based prediction of toxicity of organic compounds towards fathead minnow

Chen, Xingmei; Dang, Limin; Yang, Hai; Yu, Xinliang

doi:10.1039/d0ra05906d

Cited by 27 publications

(14 citation statements)

References 30 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…The SVM classification algorithm is based on finding the optimal separating hyperplane that is equidistant from two classes and maximizes the distance between the two parallel hyperplanes. [22][23][24]32 Given training data y 1 ; x 1 ð Þ ; . .…”

Section: Svm Classification Algorithmmentioning

confidence: 99%

“…Conversely, smaller C and c may result in under-fitting, where an SVM model has poor prediction for both the training set and the test set. [22][23][24]32 SVM parameters C and c were optimized with the GA.…”

Section: Svm Classification Algorithmmentioning

confidence: 99%

“…20,21 Many studies have confirmed that machine-learning algorithms such as support vector machine (SVM) possess better prediction ability than the multiple linear regression (MLR) method. [22][23][24] The reason for this phenomenon is attributed to the fact that MLR analysis cannot capture the nonlinear relationship between explanatory variables and the response variable.…”

mentioning

confidence: 99%

See 2 more Smart Citations

Support vector machine classification model for color fastness to ironing of vat dyes

Wang

2021

Textile Research Journal

Self Cite

View full text Add to dashboard Cite

Investigations of quantitative relationships between structure and color fastness of dyes are crucial in seeking novel dyes. For the first time, this work reported a classification model based on a quantitative structure–property relationship to predict color fastness to ironing of vat dyes. By performing binary classification analysis based on a support vector machine (SVM) and genetic algorithm, 56 vat dyes in the training set together with seven molecular descriptors were used to develop the classification model, which was validated with 59 vat dyes in the test set. The optimal SVM model ( C = 208.465 and γ = 5.9692) possesses overall accuracy of 91.1% for the training set and 83.1% for the test set, which is more accurate than those from the binary logistic regression model (87.5% and 81.4%, respectively). Furthermore, the mechanism of molecular descriptors correlated with color fastness to ironing of vat dyes is discussed.

show abstract

Section: Svm Classification Algorithmmentioning

confidence: 99%

Section: Svm Classification Algorithmmentioning

confidence: 99%

mentioning

confidence: 99%

See 1 more Smart Citation

Support vector machine classification model for color fastness to ironing of vat dyes

Wang

2021

Textile Research Journal

Self Cite

View full text Add to dashboard Cite

show abstract

“…Moreover, Chen et al developed a non-linear QSAR model to predict acute toxicity in fathead minnow species. Further, they validated it using 482 chemical compounds using a support vector machine and compared their results with other literature reported earlier [8]. Recently, Fan et al predicted toxicity compounds using different combinations of chemical descriptors and fingerprints.…”

Section: Introductionmentioning

confidence: 97%

“…Further, this strategy was time-consuming in properly splitting compounds into train and test sets [7]. In addition, developing a precise QSAR model has become difficult due to the rise in the number of samples in the test set [8]. Hence developing a successful QSAR model with high accuracy has become challenging.…”

Section: Introductionmentioning

confidence: 99%

Prediction of Micronucleus Assay Outcome Using In Vivo Activity Data and Molecular Structure Features

Ramesh

Shanthi

2021

Appl Biochem Biotechnol

View full text Add to dashboard Cite

In vivo micronucleus assay is the widely used genotoxic test to determine the extent of chromosomal aberrations caused by the chemical compounds in human beings, which plays a significant role in the drug discovery paradigm. To reduce the uncertainties of the in vivo experiments and the expenses, we intended to develop novel machine learning-based tools to predict the toxicity of the compounds with high precision. A total of 472 compounds with known toxicity information were retrieved from the PubChem Bioassay database and literature. The fingerprints and descriptors of the compounds were generated using PaDEL and ChemSAR for the analysis. The performance of the models was assessed using three tires of evaluation strategies such as 5-fold, 10-fold, and external validation. The accuracy of the models during external validation lay between 0.57 and 0.86. Note that a combination of fingerprints and random forest showed reliable predictive capability. In essence, structural alerts causing genotoxicity of the compounds were identified using the structural activity relationship model of SARpy tool. This study highlights that the structural alerts such as chlorocyclohexane and trimethylamine are likely to be the leading cause of toxicity in humans, further validated using the Toxtree application.Indeed, the results from our study will assist in scrutinizing the genotoxicity of the compounds with high precision by replacing extensive sacrifice of animal models.

show abstract

Anxiolytic and Anticonvulsant Effects of Fisetin Isolated from Bauhinia pentandra on Adult Zebrafish (Danio rerio)

da Silva,

das Chagas Lima Pinto,

Silva Marinho

et al. 2024

Chemistry & Biodiversity

View full text Add to dashboard Cite

Anxiety and epilepsy are common worldwide and represent a primary global health concern. Fisetin, a flavonoid isolated from Bauhinia pentandra, has a wide range of biological activities may be a promising alternative to combat diseases related to the central nervous system (CNS). The present study aimed to investigate the anxiolytic and anticonvulsant effects of fisetin on adult zebrafish. Furthermore, molecular docking simulations were performed to improve the results. Fisetin did not present toxicity and caused anxiolytic behavior and delayed seizures in animals. This effect may occur through serotonin neurotransmission at 5‐HT3A and/or 5‐HT3B receptors. Molecular docking simulations showed that fisetin interacts with the orthosteric site of the 5‐HT3A receptor with strong H‐bond interactions with the Trp156 residue, with a strong contribution from the catechol ring, a behavior similar to that of the antagonist co‐crystallized inhibitor granisetron (CWB). Fisetin may be a promising alternative to combat diseases related to the central nervous system. Keywords anxiety • Bauhinia pentandra • Danio rerio • epilepsy • fisetin

show abstract

Machine learning-based prediction of toxicity of organic compounds towards fathead minnow

Abstract: A quantitative structure–toxicity relationship of 963 chemicals against fathead minnow was developed by using support vector machine and genetic algorithm.

Cited by 27 publications

References 30 publications

Support vector machine classification model for color fastness to ironing of vat dyes

Support vector machine classification model for color fastness to ironing of vat dyes

Prediction of Micronucleus Assay Outcome Using In Vivo Activity Data and Molecular Structure Features

Anxiolytic and Anticonvulsant Effects of Fisetin Isolated from Bauhinia pentandra on Adult Zebrafish (Danio rerio)

Contact Info

Product

Resources

About