Predicting the biological activities of triazole derivatives as SGLT2 inhibitors using multilayer perceptron neural network, support vector machine, and projection pursuit regression models

Yuan, Jun; Yu, Shuling; Gao, Shanxing; Gan, Ying; Zhang, Yi; Zhang, Ting; Wang, Yali; Liu, Yang; Shi, Jiahua; Wu, Yao

doi:10.1016/j.chemolab.2016.06.002

Cited by 10 publications

(3 citation statements)

References 42 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…CATS2D_02_AL is a descriptor that belongs to the chemically advanced template search-2D (CATS2D) descriptor category, which deals with acceptor-lipophilic at lag 02 . In other words, this descriptor defines hydrogen-bond acceptor (A) and lipophilic (L) points at topological distances of 2 bonds . Increasing the number of these fragments will significantly increase the compound’s toxicity.…”

Section: Resultsmentioning

confidence: 99%

“…90 In other words, this descriptor defines hydrogen-bond acceptor (A) and lipophilic (L) points at topological distances of 2 bonds. 91 Increasing the number of these fragments will significantly increase the compound's toxicity. nCH2RX_C1_assay is a fragmental descriptor representing the number of carbon atoms adjacent to a halogen or reactive functional group.…”

Section: Illustrates How This Technique Workmentioning

confidence: 99%

See 1 more Smart Citation

Multi-Endpoint Acute Toxicity Assessment of Organic Compounds Using Large-Scale Machine Learning Modeling

Daghighi,

Casanola-Martin,

Iduoku

et al. 2024

Environ. Sci. Technol.

View full text Add to dashboard Cite

In recent years, alternative animal testing methods such as computational and machine learning approaches have become increasingly crucial for toxicity testing. However, the complexity and scarcity of available biomedical data challenge the development of predictive models. Combining nonlinear machine learning together with multicondition descriptors offers a solution for using data from various assays to create a robust model. This work applies multicondition descriptors (MCDs) to develop a QSTR (Quantitative Structure–Toxicity Relationship) model based on a large toxicity data set comprising more than 80,000 compounds and 59 different end points (122,572 data points). The prediction capabilities of developed single-task multi-end point machine learning models as well as a novel data analysis approach with the use of Convolutional Neural Networks (CNN) are discussed. The results show that using MCDs significantly improves the model and using them with CNN-1D yields the best result (R 2 train = 0.93, R 2 ext = 0.70). Several structural features showed a high level of contribution to the toxicity, including van der Waals surface area (VSA), number of nitrogen-containing fragments (nN+), presence of S–P fragments, ionization potential, and presence of C–N fragments. The developed models can be very useful tools to predict the toxicity of various compounds under different conditions, enabling quick toxicity assessment of new compounds.

show abstract

Section: Resultsmentioning

confidence: 99%

Section: Illustrates How This Technique Workmentioning

confidence: 99%

Multi-Endpoint Acute Toxicity Assessment of Organic Compounds Using Large-Scale Machine Learning Modeling

Daghighi,

Casanola-Martin,

Iduoku

et al. 2024

Environ. Sci. Technol.

View full text Add to dashboard Cite

show abstract

“…The machine learning models find relations between inputs and outputs even if the representation is impossible; this characteristic allows the use of machine learning models in the case of forecasting problems as described in the previous report [32] . Support vector machines (SVMs) are a type of novel machine learning method based on statistical learning theory, which is a new powerful tool for identification and prediction in nonlinear systems [33] . Particle swarm optimization (PSO) is generally used with SVM for feature selection and extraction [34,35] , parameter optimization of SVM [36][37][38][39][40] and multi-class classification in SVM [41] .…”

Section: Introductionmentioning

confidence: 99%

PSO-SVM applied to SWASV studies for accurate detection of Cd(II) based on disposable electrode

Zhao

Wang

Yin

et al. 2017

International Journal of Agricultural and Biological Engineering

View full text Add to dashboard Cite

Square wave anodic stripping voltammetry (SWASV) is an effective method for the detection of Cd(II), but the presence of Pb(II) usually has some potential and negative interference on the SWASV detection of Cd(II). In this paper, a novel method was proposed to predict the concentration of Cd(II) in the presence of Pb(II) based on the combination of chemically modified electrode (CME), machine learning algorithms (MLA) and SWASV. A Bi film/ionic liquid/screenprinted electrode (Bi/IL/SPE) was prepared and used for the sensitive detection of trace Cd(II). The parameters affecting the stripping currents were investigated and optimized. The morphologies and electrochemical properties of the modified electrode were characterized by scanning electron microscopy (SEM) and SWASV. The measured SWASV spectrograms obtained at different concentrations were used to build the mathematical models for the prediction of Cd(II), which taking the combined effect of Cd(II) and Pb(II) into consideration on the SWASV detection of Cd(II), and to establish a nonlinear relationship between the stripping currents of Pb(II) and Cd(II) and the concentration of Cd(II). The proposed mathematical models rely on an improved particle swarm optimization-support vector machine (PSO-SVM) to assess the concentration of Cd(II) in the presence of Pb(II) in a wide range of concentrations. The experimental results suggest that this method is suitable to fulfill the goal of Cd(II) detection in the presence of Pb(II) (correlation coefficient, mean absolute error and root mean square error were 0.998, 1.63 and 1.68, respectively). Finally, the proposed method was applied to predict the trace Cd(II) in soil samples with satisfactory results.

show abstract

Drug design by machine-trained elastic networks: predicting Ser/Thr-protein kinase inhibitors’ activities

2020

View full text Add to dashboard Cite

Predicting the biological activities of triazole derivatives as SGLT2 inhibitors using multilayer perceptron neural network, support vector machine, and projection pursuit regression models

Cited by 10 publications

References 42 publications

Multi-Endpoint Acute Toxicity Assessment of Organic Compounds Using Large-Scale Machine Learning Modeling

Multi-Endpoint Acute Toxicity Assessment of Organic Compounds Using Large-Scale Machine Learning Modeling

PSO-SVM applied to SWASV studies for accurate detection of Cd(II) based on disposable electrode

Drug design by machine-trained elastic networks: predicting Ser/Thr-protein kinase inhibitors’ activities

Contact Info

Product

Resources

About