Prediction of molecular toxicity plays an important role in drug discovery. It has a direct relationship with human health and medical destiny. Accurately assessing a molecule’s toxicity can aid in the weeding out of low-quality com pounds early in the drug discovery phase, avoiding depletion later in the drug development process. Computational models have been used to automatically predict the molecular toxicity. In this paper, a machine learning-based model has been proposed. TF/IDF representation scheme has been used for N-gram and integrated with SMILE. Multiple machine learning classifiers have been im plemented such as Decision Tree, KNN, MLP, Random Forests, SGD and SVM. RF outperforms all the other classifiers. A wide range of N-gram models have been implemented and trigram reported the best results.
1 May 2025 in International Journal of Applied Intelligent Computing and Informatics
The increasing number of chemicals has aroused public concern due to their negative influence on the environment and human health. To protect the environment and human health, the toxicity of these compounds must be assessed. Traditional in vitro and in vivo toxicity testing are time-consuming, expensive, and complex, and they may pose ethical considerations as well. Due to these restrictions, alternative methods for assessing the toxicity of a chemical are required. Numerous toxicity prediction models have been developed recently using a variety of machine learning and deep learning algorithms such as support vector machines, random forests, k-nearest neighbors, ensemble learning, and deep neural networks by integrating classical ML techniques or Deep Learning (DL) with molecular representations such as fingerprints or 2D graphs. This paper presents an overview of chemical toxicity and the drug Discovery Process. It summarizes current ML and DL models for predictive toxicology with a brief objective and the limitations and challenges AI faces in toxicity prediction.