In Silico Prediction of Chemical Acute Oral Toxicity Using Multi-Classification Methods

Abstract

Chemical acute oral toxicity is an important end point in drug design and environmental risk assessment. However, it is difficult to determine by experiments, and in silico methods are hence developed as an alternative. In this study, a comprehensive data set containing 12, 204 diverse compounds with median lethal dose (LD₅₀) was compiled. These chemicals were classified into four categories, namely categories I, II, III and IV, based on the criterion of the U.S. Environmental Protection Agency (EPA). Then several multiclassification models were developed using five machine learning methods, including support vector machine (SVM), C4.5 decision tree (C4.5), random forest (RF), κ-nearest neighbor (kNN), and naïve Bayes (NB) algorithms, along with MACCS and FP4 fingerprints. One-against-one (OAO) and binary tree (BT) strategies were employed for SVM multiclassification. Performances were measured by two external validation sets containing 1678 and 375 chemicals, separately. The overall accuracy of the MACCS-SVM(OAO) model was 83.0% and 89.9% for external validation sets I and II, respectively, which showed reliable predictive accuracy for each class. In addition, some representative substructures responsible for acute oral toxicity were identified using information gain and substructure frequency analysis methods, which might be very helpful for further study to avoid the toxicity.

Tetrahydrobenzo[h][1,6]naphthyridine-6-chlorotacrine hybrids as a new family of anti-Alzheimer agents targeting β-amyloid, tau, and cholinesterase pathologies

“…Of note, the predicted in vitro ability of the novel hybrids to cross the BBB was also confirmed through the BBB permeation index obtained using a recently reported in silico multiclassification method (Table 2), which was developed utilizing a comprehensive data set containing around 12000 diverse compounds [29]. This method was also used to assess the intestinal absorption of the novel compounds, which was predicted to be positive in all cases.…”

Synthesis, Pharmacological Evaluation, and Docking Studies of Novel Pyridazinone‐Based Cannabinoid Receptor Type 2 Ligands

“…Therefore, it is important to predict whether an ew compound is as ubstrate, an inhibitor or an inducer of these CYP isoforms in early drug discoveryp hases. [39,40] Therefore, takingt he pharmacological and ADMET prediction data into consideration, the highly selectiveC B 2 Ri nverse agonist 22 might serve as al ead structure for furthero ptimizing this novel class of CB 2 Rl igandsi nt erms of potencya nd pharmacokinetic properties. This differencei nt he metabolism among pyridazinonesm ight be due to the presenceo fh alogens in the molecule (derivatives 22-24 have only one halogen,w hereas the other derivatives have at least two halogensper structure).…”

Molecular similarity-based predictions of the Tox21 screening outcome

“…ECFP4 is a member of the extended-connectivity fingerprint type often used to analyze structure-activity relationships of small molecules (Rogers and Hahn, 2010). MACCS keys are another frequently used fingerprint type which encodes the presence of specific substructures and has been successfully used for predictions of acute oral toxicity (Li et al, 2014). The ToxPrint fingerprint (Yang et al, 2015a) is based on a library of more than 700 chemotypes which represent molecules in public chemical and toxicity databases and cover substructures associated with toxic effects and thus may be of particular importance for in silico toxicity predictions.…”

“…models, support vector machines, random forests or ensembles of different classification methods can use the similarity defined the molecular structure and properties to make predictions for novel compounds. This concept has also been frequently and successfully applied to predictions of various toxicological endpoints (Drwal et al, 2014;Gadaleta et al, 2014;Li et al, 2014;Liu et al, 2015).…”