Abstract
The use of in silico toxicity prediction methods plays an important role in the selection of lead compounds and in ADMET studies since in vitro and in vivo methods are often limited by ethics, time, budget and other resources. In this context, we present our new web tool VenomPred, a user-friendly platform for evaluating the potential mutagenic, hepatotoxic, carcinogenic and estrogenic effects of small molecules. VenomPred platform employs several in-house Machine Learning (ML) models developed with datasets derived from VEGA QSAR, a software that includes a comprehensive collection of different toxicity models and has been used as a reference for building and evaluating our ML models. The results showed that our models achieved equal or better performance than those obtained with the reference models included in VEGA QSAR. In order to improve the predictive performance of our platform, we adopted a consensus approach combining the results of different ML models, which was able to predict chemical toxicity better than the single models. This improved method was thus implemented in the VenomPred platform, a freely accessible webserver that takes the SMILES (Simplified Molecular-Input Line-Entry System) strings of the compounds as input and sends the prediction results providing a probability score about their potential toxicity.
Highlights
The use of artificial intelligence (AI) and, in particular, machine learning (ML) in toxicology is increasingly widespread due to the reduced costs and time required by in silico approaches compared to in vivo and in vitro studies, the ethical concerns related to animal experiments and the possibility to handle and process a large amount of data through ML models
Four hyperparameters were tuned in order to minimize the error in the path from the input to the output predictions: (a) hidden_layer_size, which determines the number of neurons and the number of hidden layers; (b) solver, which is fundamental to optimize the predictions at every decision step through the different layers; (c) activation, which refers to the activation function and defines how the weighted sum of the input is transformed into output by one or more nodes in a network layer; (d) learning_rate_init, which controls the step-size in updating the weights
With the aim of developing ML models for predicting the potential toxicological effect of small molecules and comparing their performance with known toxicity models reported in the literature, we used VEGA QSAR software as a reference
Summary
The use of artificial intelligence (AI) and, in particular, machine learning (ML) in toxicology is increasingly widespread due to the reduced costs and time required by in silico approaches compared to in vivo and in vitro studies, the ethical concerns related to animal experiments and the possibility to handle and process a large amount of data through ML models. VEGAHub provides the freely available software VEGA QSAR, able to perform toxicity predictions employing in silico methods and models reported in the literature and implemented therein. After developing multiple models focused on the four different endpoints, in order to evaluate the possibility of improving their predictive performance, a consensus approach combining the results of their predictions was applied. As this approach was able to predict chemical toxicity better than the single models, it was implemented in the VenomPred platform, a freely accessible web tool that takes the SMILES strings of the compounds as an input, providing a probability score about their potential toxicity as an output
Sign-in/Register to view highlights & summary Talk to us
Join us for a 30 min session where you can share your feedback and ask us any queries you have
Schedule a callDisclaimer: All third-party content on this website/platform is and will remain the property of their respective owners and is provided on "as is" basis without any warranties, express or implied. Use of third-party content does not indicate any affiliation, sponsorship with or endorsement by them. Any references to third-party content is to identify the corresponding services and shall be considered fair use under The CopyrightLaw.
