Deep Learning Based Regression and Multiclass Models for Acute Oral Toxicity Prediction with Automatic Chemical Feature Extraction

Abstract

Median lethal death, LD, is a general indicator of compound acute oral toxicity (AOT). Various in silico methods were developed for AOT prediction to reduce costs and time. In this study, we developed an improved molecular graph encoding convolutional neural networks (MGE-CNN) architecture to construct three types of high-quality AOT models: regression model (deepAOT-R), multiclassification model (deepAOT-C), and multitask model (deepAOT-CR). These predictive models highly outperformed previously reported models. For the two external data sets containing 1673 (test set I) and 375 (test set II) compounds, the R and mean absolute errors (MAEs) of deepAOT-R on the test set I were 0.864 and 0.195, and the prediction accuracies of deepAOT-C were 95.5% and 96.3% on test sets I and II, respectively. The two external prediction accuracies of deepAOT-CR are 95.0% and 94.1%, while the R and MAE are 0.861 and 0.204 for test set I, respectively. We then performed forward and backward exploration of deepAOT models for deep fingerprints, which could support shallow machine learning methods more efficiently than traditional fingerprints or descriptors. We further performed automatic feature learning, a key essence of deep learning, to map the corresponding activation values into fragment space and derive AOT-related chemical substructures by reverse mining of the features. Our deep learning architecture for AOT is generally applicable in predicting and exploring other toxicity or property end points of chemical compounds. The two deepAOT models are freely available at http://repharma.pku.edu.cn/DLAOT/DLAOThome.php or http://www.pkumdl.cn/DLAOT/DLAOThome.php .

Machine Learning Based Toxicity Prediction: From Chemical Structural Description to Transcriptome Analysis

“…This rapid development is partly stimulated by its many important applications, one of which is drug toxicity prediction in silico [88,127,158]. Together with “Big Data” science [159], machine learning techniques may provide much more information about toxicity than ever before.…”

“…The fingerprint was further mined both forwardly and backwardly, which yielded the deep-minded fingerprint (the array of black dots in Figure 3). The deep-minded fingerprint was then tested by the regression model (the blue circle) and the multiclass/multitask models (the green circles), which yielded a classification accuracy up to 95.0% and a regression R square value of 0.861 [127]. …”

“…All of the pre-fingerprints are integrated to generate the fingerprint, which was further processed to generate the deep-mined fingerprint. The deep-minded fingerprint was then tested by the regression model (the blue circle) and the multiclass/multitask models (the green circles) [127]. …”

DeepNose: Using artificial neural networks to represent the space of odorants

“…In this study, we use multilayer (deep) convolutional neural networks (CNNs) 6 use derived representations to predict perceptual responses neural networks that use ensembles of spatial filters of increasing complexity to to representing objects present in the input Networks based on CNNs are the best performing method for image recognition better at classifying images than performance of CNNs in 2D suggests that they are well types of spatial data, such as the 3D molecular structures. Deep networks have indeed been applied to classify molecules, including the predictions of pharmacokinetics properties, toxicity, and protein ligand interaction [10][11][12] DeepNose: Using artificial neural networks to represent the space of Tran, Daniel Kepple, Sergey Cold Spring Harbor Laboratory, Cold system employs an ensemble of to derive olfactory percepts. W and used that representation to predict human hypothesized that ORs may be considered trained using conventional machine learning called DeepNose, to deduce a low represented by their 3D spatial structure predicting physical properties and odorant percepts found that despite the lack of human expertise, DeepNose features led to perceptual predictions of comparable accuracy to molecular DeepNose network can extract can help understand…”

NeoDTI: Neural integration of neighbor information from a heterogeneous network for discovering new drug-target interactions

“…For instance, the DTI network data have been integrated with the drug structure and protein sequence information into a networkbased machine learning model (e.g., a regularized least squares framework) for predicting new DTIs (Xia et al, 2010;van Laarhoven et al, 2011;van Laarhoven and Marchiori, 2013). Inspired by the recent surge of deep learning techniques, models with higher predictive capacity have also been developed in various drug discovery settings (e.g., compound-protein interaction prediction, drug discovery with one-shot learning) (Wang and Zeng, 2013;Wan and Zeng, 2016;Hamanaka et al, 2017;Tian et al, 2016;Altae-Tran et al, 2017;Xu et al, 2017).…”