Prediction of human fetal–maternal blood concentration ratio of chemicals

Abstract

Background The measurement of human fetal-maternal blood concentration ratio (logFM) of chemicals is critical for the risk assessment of chemical-induced developmental toxicity. While a few in vitro and ex vivo experimental methods were developed for predicting logFM of chemicals, the obtained experimental results are not able to directly predict in vivo outcomes. Methods A total of 55 chemicals with logFM values representing in vivo fetal-maternal blood ratio were divided into training and test datasets. An interpretable linear regression model was developed along with feature selection methods. Cross-validation on training dataset and prediction on independent test dataset were conducted to validate the prediction model. Results This study presents the first valid quantitative structure-activity relationship model following the Organisation for Economic Co-operation and Development (OECD) guidelines based on multiple linear regression for predicting in vivo logFM values. The autocorrelation descriptor AATSC1c and information content descriptor ZMIC1 were identified as informative features for predicting logFM. After the adjustment of the applicability domain, the developed model performs well with correlation coefficients of 0.875, 0.850 and 0.847 for model fitting, leave-one-out cross-validation and independent test, respectively. The model is expected to be useful for assessing human transplacental exposure.

The Role of the 3Rs for Understanding and Modeling the Human Placenta

“…These models correlate the structural and chemical properties of the substances with the available biological data associated with their permeability/clearance across the placenta [ 98 , 99 ]. The biological data are often obtained from results of ex vivo experiments reported in the literature [ 82 , 100 ] or from concentrations of compounds found in maternal blood and umbilical cord [ 99 , 101 , 102 , 103 , 104 ]. These QSAR models have analyzed the placental transfer of diverse substances such as persistent organic pollutants [ 102 ], dioxins [ 102 ], widespread chemicals [ 82 , 103 , 104 ], therapeutic drugs [ 82 , 99 , 103 , 104 ] and even compounds from tocolytic herbs [ 101 ].…”

“…The biological data are often obtained from results of ex vivo experiments reported in the literature [ 82 , 100 ] or from concentrations of compounds found in maternal blood and umbilical cord [ 99 , 101 , 102 , 103 , 104 ]. These QSAR models have analyzed the placental transfer of diverse substances such as persistent organic pollutants [ 102 ], dioxins [ 102 ], widespread chemicals [ 82 , 103 , 104 ], therapeutic drugs [ 82 , 99 , 103 , 104 ] and even compounds from tocolytic herbs [ 101 ]. As an output of the statistic-derived analysis, the models have also identified potential molecular characteristics of the different substances that can be determinants for placental transfer.…”

Ensemble learning for predicting ex vivo human placental barrier permeability

“…Due to the complexity and ethical issues, the measurement of placental transfer permeability of chemicals based on in vivo conditions may not be applicable for risk assessment of novel chemicals. Therefore, a few computational methods have been developed to predict in vivo human fetal-maternal blood concentration ratio (logFM) of chemicals [1,2]. Takaku et al conducted the first work on the prediction model for predicting in vivo logFM values based on manually curated 55 chemicals with logFM values [2].…”

“…Takaku et al conducted the first work on the prediction model for predicting in vivo logFM values based on manually curated 55 chemicals with logFM values [2]. Subsequently, Wang et al proposed the first valid model according to the OECD guideline with fewer features but higher performance [1].…”

A Machine Learning Model to Predict Drug Transfer Across the Human Placenta Barrier

“…Takaku et al Takaku et al, (2015) developed a QSAR model for predicting the in vivo fetal–maternal blood concentration ratio (F/M ratio) for a set of 55 compounds. Later, Wang et al, using the same chemical library of 55 compounds as Takaku et al Takaku et al, (2015) , developed a QSAR model following the Organization for Economic Co-operation and Development (OECD) guidelines based on multiple linear regression adjustments for predicting in vivo log (F/M) values ( Wang et al, 2020 ). These studies achieved a reasonable predictive potential (the correlation between measured and predicted values is acceptable); however, all of them were validated with few samples.…”