Research on a Prediction Model for the Bioconcentration Factor (BCF) of Polyhalogenated Organic Phosphates Based on QSPR

The bioconcentration factor (BCF) is a key parameter for evaluating the environmental behavior and ecological risks of organic pollutants, yet experimental determination of BCF values is time-consuming and unsuitable for large-scale screening. Polyhalogenated organophosphate esters, widely used as flame retardants, have raised increasing environmental concern due to their persistence and potential bioaccumulation. In this study, a quantitative structure–property relationship (QSPR) modeling framework was developed to predict the BCF of polyhalogenated organophosphate esters and support environmental risk prioritization. A dataset consisting of 160 compounds was compiled and divided into training (n = 130) and test (n = 30) sets. Ten informative molecular descriptors were selected from an initial pool of 766 candidates using a genetic algorithm. Three predictive models, including multiple linear regression (MLR), support vector machine (SVM), and backpropagation artificial neural network (BP-ANN), were constructed and systematically evaluated. Dataset partitioning was verified using Tanimoto similarity analysis and uniform manifold approximation and projection (UMAP) visualization to ensure structural independence and chemical-space representativeness. Model performance was assessed through a comprehensive validation scheme employing 11 statistical metrics covering internal validation, external predictivity, and robustness. The applicability domain was further defined using Williams plots to identify reliable prediction boundaries and outlier compounds. Among the three models, the BP-ANN exhibited the best overall predictive performance, achieving R ² _train = 0.86 and Q ² _F2 = 0.79, with low external prediction errors ( MAE _test = 0.33 and RMSE _test = 0.41). The results indicate that bioaccumulation behavior of polyhalogenated organophosphate esters is jointly governed by hydrophobicity, molecular topology, electronic distribution, and functional group composition. The proposed QSPR framework provides a reliable and efficient tool for screening the bioaccumulation potential of structurally related flame retardants and may assist in early-stage environmental risk assessment and sustainable chemical management.