Polycyclic Aromatic Hydrocarbons (PAHs) are a complex group of environmental contaminants with long environmental half-lives. As these compounds slowly degrade, numerous intermediate structures are created. Previous research suggests that the resulting changes in structure and functional groups can result in a significant increase in mutagenicity over the original PAHs. Each individual PAH can potentially degrade into several thousand unique transformation products, creating a complex mixture of intermediates. The microbial degradation process in the environment results in mutagenic activation similar to the metabolic activation that can occur in an organism. A combination of structural and energetic descriptors has proven effective in similar studies for classifying which compounds will be mutagenic. However, most existing studies of environmental PAH mutagens primarily focus on nitrogenated derivatives, which are prevalent in the atmosphere, but are not as relevant in soil, where PAHs spend the vast majority of their environmental life. Additionally, PAHs commonly found in the environment can range from as large as five rings to as small as a single ring, requiring a broach-reaching methodology to evaluating their mutagenic potential. In this presentation we describe a newly developed combination of supervised and unsupervised machine learning methods which provides a solution to identifying the mutagenic intermediates with improved performance over the currently available tools in this field. K-means clustering with Principal Component Analysis allows us to identify molecule clusters that have similar mechanisms of action and recursive feature elimination identifies the most applicable descriptors. The chemical descriptors identified by our analysis suggests that the clusters have different primary biological mechanisms of mutagenicity, as well as identifying common structures that enhance or reduce the likelihood of mutagenicity. This cluster-specific method out-performs available classifiers in predicting direct acting mutagens resulting from the biodegradation of PAHs, and provides insight into future studies evaluating the environmental hazards resulting from PAH biodegradation.

Cluster-based classifier to identify direct-acting mutagenic PAH transformation products

ACS Spring Meeting 2021

The official archive and repository for all ACS Meetings & Expositions abstracts and select posters or presentations.