Predicting Placental Transfer of Environmental Chemicals Incorporating QSAR and Q-RASAR Modeling Approaches
Brayan Martinez
Co-Presenters: Individual Presentation
College: The Dorothy and George Hennings College of Science, Mathematics and Technology
Major: Computer Information Systems (M.S.)
Faculty Research Mentor: Supratik Kar
Abstract:
Humans are exposed to a diverse chemicals daily through food, beverages, and consumer products. While many are harmless, some pose potential risks, particularly during critical developmental periods such as pregnancy. Prenatal exposure to certain chemicals has been linked to adverse fetal development and long-term health consequences. This study develops a computational approach to assess and predict chemical transfer from mother to fetus, quantified by the cord-to-mother serum (CS:MS) concentration ratio, using chemical structural features. The analysis began with a dataset of 105 environmental chemicals, each with a reported CS:MS ratio, transformed into ln(CS:MS) values for modeling. A Quantitative Structure-Activity Relationship (QSAR) model was first employed to identify key physicochemical and structural features influencing CS:MS ratios. Building on this, a quantitative read-across structure-activity relationship (q-RASAR) model was developed, incorporating similarity features derived from read-across methodologies. The q-RASAR model demonstrated superior predictive performance over the QSAR model based on validation criteria. The final q-RASAR model was applied to estimate CS:MS ratios for untested environmental chemicals, providing critical insights into fetal exposure to potentially harmful substances. This computational tool helps address data gaps in prenatal chemical toxicity and supports the development of strategies to mitigate exposure risks.