Computational characterisation of ecological hazard and risk of environmental mixtures

The project Computational characterisation of ecological hazard and risk of environmental mixtures focused on two key priorities in advancing a next-generation risk assessment (NGRA) workflow: (i) transitioning to assessing mixtures and (ii) utilising mechanism-based hazard assessment based on new approach methodology (NAM) data. Through a case study, an enhanced component-based mixture risk assessment (CBMRA) framework, integrating high-throughput-screening (HTS) bioactivity data combined with a quantitative adverse outcome pathway (qAOP) approach for hazard and risk assessment was demonstrated. The case study utilised a previously published qAOP based on the proposed AOP-Wiki AOP#245 'Uncoupling of photophosphorylation leading to reduced ATP production associated growth inhibition' (Xie et al., 2018), which models growth inhibition via uncoupling of mitochondrial oxidative phosphorylation from a reference chemical on a model aquatic plant (Moe et al., 2021). It used pesticide monitoring data from the European Environmental Agency's Pesticide Indicator dataset (WISE statistics - Pesticides) in freshwater environments (European Environment Agency, 2023). Relevant bioactivity data from ToxCast and Tox21 were mapped to the target pesticides to derive equipotent mixture compared to the reference chemical used to parametrised the qAOP (i.e. 3,5-dichlorophenol), enabling assessment of the potential initiation of the AOP cascade. In silico methods were used to fill bioactivity data gaps and probabilistic modelling using a Bayesian network (BN) was designed to incorporate various uncertainties into the current NGRA workflow. The study assessed the strengths and limitations of a NAM-based CBMRA, with a particular attention on equipotency assessment as a means to extend the chemical domain of qAOPs for mechanistic ecological risk assessment. It highlighted both conceptual and technical innovations and identified research needs to improve the approach towards future regulatory adoption.