PREMIER partners publish new scientific paper: Intelligent consensus predictions of bioconcentration factor of pharmaceuticals using 2D and fragment-based descriptors

As part of the PREMIER project, we aim to evaluate the largely unknown environmental impact of pharmaceuticals by using new approach methodologies (NAMs), such as in-silico methods. This strategy reduces reliance on animal experiments while providing regulators with valuable information to support the prioritization of compounds for further assessment. Moreover, it can save time by allowing the identification of the most problematic chemicals and directing them for more in-depth testing.

In this context, the present study aims to evaluate the ecotoxicity potential of 122 pharmaceutical compounds and to identify key structural features influencing their behaviour, using the bioconcentration factor (BCF) in fish as the main indicator. To this end, a regression-based quantitative structure–property relationship (QSPR) approach was applied to determine the chemical characteristics responsible for acute bioconcentration in fish. Multiple models were developed and combined using an intelligent consensus algorithm, allowing an improvement in the overall predictive performance of the regression-based models.

The results indicate that the presence of large, lipophilic, and electronegative molecular groups significantly increases the bioaccumulation potential of pharmaceuticals, whereas hydrophilic properties show a negative influence on BCF values. The information obtained from the modelled descriptors may be useful for future aquatic environmental risk assessments. Moreover, the proposed modelling framework can act as an early warning tool to identify pharmaceuticals with a higher likelihood of causing adverse effects in aquatic ecosystems, thereby supporting regulatory evaluation and decision-making processes

See the article here.