A review on machine learning methods for<i>in silico</i>toxicity prediction

Idakwo, Gabriel; Luttrell, Joseph; Chen, Minjun; Hong, Huasheng; Zhou, Zhaoxian; Gong, Ping; Zhang, Chaoyang

doi:10.1080/10590501.2018.1537118

Cited by 107 publications

(78 citation statements)

References 77 publications

Supporting

Mentioning

Contrasting

Unclassified

Order By: Relevance

“…The support vector machines (SVM) algorithm employs a range of kernel functions (e.g., linear, polynomial, radial, etc.) to maximize the decision boundary between classes and to define a hyperplane able to best discriminate the classes [96]. It is an algorithm apt for dealing with a large number of features and has been used with good results to solve a diverse range of classification and regression tasks, including QSAR investigations [97,98].…”

Section: Classification Algorithmsmentioning

confidence: 99%

Computational Models Using Multiple Machine Learning Algorithms for Predicting Drug Hepatotoxicity with the DILIrank Dataset

Ancuceanu

Hovaneț

Anghel

et al. 2020

IJMS

View full text Add to dashboard Cite

Drug-induced liver injury (DILI) remains one of the challenges in the safety profile of both authorized and candidate drugs, and predicting hepatotoxicity from the chemical structure of a substance remains a task worth pursuing. Such an approach is coherent with the current tendency for replacing non-clinical tests with in vitro or in silico alternatives. In 2016, a group of researchers from the FDA published an improved annotated list of drugs with respect to their DILI risk, constituting “the largest reference drug list ranked by the risk for developing drug-induced liver injury in humans” (DILIrank). This paper is one of the few attempting to predict liver toxicity using the DILIrank dataset. Molecular descriptors were computed with the Dragon 7.0 software, and a variety of feature selection and machine learning algorithms were implemented in the R computing environment. Nested (double) cross-validation was used to externally validate the models selected. A total of 78 models with reasonable performance were selected and stacked through several approaches, including the building of multiple meta-models. The performance of the stacked models was slightly superior to other models published. The models were applied in a virtual screening exercise on over 100,000 compounds from the ZINC database and about 20% of them were predicted to be non-hepatotoxic.

show abstract

Section: Classification Algorithmsmentioning

confidence: 99%

Computational Models Using Multiple Machine Learning Algorithms for Predicting Drug Hepatotoxicity with the DILIrank Dataset

Ancuceanu

Hovaneț

Anghel

et al. 2020

IJMS

View full text Add to dashboard Cite

show abstract

“…It uses a variety of kernel functions (e.g. linear, polynomial, radial etc) to project features in a vector space maximizing the partitioning boundary between classes and to identify the hyperplane that best discriminates the classes (41).…”

Section: Machine Learning Algorithms and Model Buildingmentioning

confidence: 99%

Use of QSAR Global Models and Molecular Docking for Developing New Inhibitors of c-Src Tyrosine Kinase

Ancuceanu

Tamba

Stoicescu³

et al. 2019

Preprint

View full text Add to dashboard Cite

Prototype of a family of at least nine members, c-src tyrosine kinase is a therapeutically interesting target, because its inhibition might be of interest not only in a number of malignancies, but also in a diverse array of conditions, from neurodegenerative pathologies to certain viral infections. Computational methods in drug discovery are considerably cheaper than conventional methods and offer opportunities of screening very large numbers of compounds in conditions that would be simply impossible within the wet lab experimental settings. We have explored the use of global QSAR models and molecular ligand docking in the discovery of new c-src tyrosine kinase inhibitors. Using a data set of 1038 compounds from ChEMBL and 19 blocks of molecular descriptors, we have developed over 200 QSAR classification models, based on six machine learning algorithms and 17 feature selection methods. We have selected 49 with reasonably good performance (positive predictive value and balanced accuracy higher than 70% in nested cross validation) and the models were assembled by stacking with a simple majority vote and used for the virtual screening of over the “named” ZINC data set (over 100,000 compounds). 744 compounds were predicted by at least 50% of the QSAR models as active, 147 compounds were within the applicability domain and predicted by at least 75% of the models to be active. The latter 147 compounds were submitted to molecular ligand docking using Vina and Ledock, and a number of 90 were predicted to be active based on the binding energy. External data from CHEMBL and PUBCHEM confirmed that at least 7.83% (in the case of QSAR) or 6.67% (in the case of integrated QSAR and molecular docking) of the compounds are active on the c-src target.

show abstract

“…Quantitative structure-activity relationship, or QSAR (Figure 1), is an area of molecular modeling that studies relationships between structure and activity using mathematical statistics and machine learning methods. QSAR is efficiently used to predict toxicity of chemical substances [9][10][11][12][13]. Classical QSAR is a so-called Hansch analysis [14], which stands on the assumption that bioactivity of compounds is correlated with geometrical and physicochemical descriptors.…”

Section: Introductionmentioning

confidence: 99%

Nano-(Q)SAR for Cytotoxicity Prediction of Engineered Nanomaterials

2019

View full text Add to dashboard Cite

Although nanotechnology is a new and rapidly growing area of science, the impact of nanomaterials on living organisms is unknown in many aspects. In this regard, it is extremely important to perform toxicological tests, but complete characterization of all varying preparations is extremely laborious. The computational technique called quantitative structure-activity relationship, or QSAR, allows reducing the cost of time-and resource-consuming nanotoxicity tests. In this review, (Q)SAR cytotoxicity studies of the past decade are systematically considered. We regard here five classes of engineered nanomaterials (ENMs): Metal oxides, metal-containing nanoparticles, multi-walled carbon nanotubes, fullerenes, and silica nanoparticles. Some studies reveal that QSAR models are better than classification SAR models, while other reports conclude that SAR is more precise than QSAR. The quasi-QSAR method appears to be the most promising tool, as it allows accurately taking experimental conditions into account. However, experimental artifacts are a major concern in this case.

show abstract

A review on machine learning methods forin silicotoxicity prediction

Cited by 107 publications

References 77 publications

Computational Models Using Multiple Machine Learning Algorithms for Predicting Drug Hepatotoxicity with the DILIrank Dataset

Computational Models Using Multiple Machine Learning Algorithms for Predicting Drug Hepatotoxicity with the DILIrank Dataset

Use of QSAR Global Models and Molecular Docking for Developing New Inhibitors of c-Src Tyrosine Kinase

Nano-(Q)SAR for Cytotoxicity Prediction of Engineered Nanomaterials

Contact Info

Product

Resources

About