Modeling the transplacental transfer of small molecules: A case study on poly/perfluorinated substances (PFAS)

Background: Despite their large numbers and widespread use, very little is known about the extent to which per- and polyfluoroalkyl substances (PFAS) are transferred to the fetus during pregnancy. Objective: The aim of our study is to develop computational approaches that can be used to evaluate the extent to which PFAS can cross to cross the placenta and partition to cord blood. Methods: We collected experimental values of the central tendency of concentration ratio between cord and maternal blood (RCM) for 264 chemical compounds and calculated their physicochemical descriptors. We developed and tested three machine learning models, an artificial neural network (ANN), a random forest (RF), and a support vector machine (SVM), and used the compiled database to train the models. We then applied our best-performing models to make predictions of RCM for a large dataset of PFAS chemicals (n=4,292). We, finally, used the calculated descriptors of the chemicals to identify which properties correlated significantly with RCM. Results: ANN showed the best performance in terms of accuracy and predictive power. Our predictions of RCM for PFAS suggest that 2,030 compounds are expected to partition favorably to cord blood. Significance: This observation has public important implications as many PFAS have been shown to interfere with fetal development