Recent advancements in QSAR and machine learning approaches for risk assessment of organic chemicals

The chemical risk assessment process, which serves as the foundation for controlling the hazards of chemical exposure, has a significant influence on the economy, the health of hundreds of millions of people, and the state of the environment. In this advanced industrialized period, all living creatures and the environment are exposed to many kinds of chemicals such as organics, pesticides, heavy metals, and medications, which may present direct or indirect risks to humans, wildlife, aquatic systems, and ecosystems. However, due to a shortage of resources for testing, interference from third-party interests, and the sheer volume of potentially relevant information on the chemicals from various sources, the toxicity data for individual organic chemicals is accessible for a subset of all chemicals found in the environment. As a result, for determining the risk of dangerous substances or chemicals, there has been a growth in the use of nonanimal methodologies, such as in silico and/or in vitro techniques. Several statistical and machine-learning techniques relate the biological activity of molecules to molecular descriptors that interpret chemical structure features. This chapter focuses on statistical and machine learning methodologies for predicting the toxicity of organic chemicals and their latest applications in the toxicity prediction of organic chemicals. This chapter also discusses the fundamentals of machine learning algorithms and their applications in predictive toxicology, which may be helpful for researchers of computational toxicology.