Computational approaches to evaluate ecotoxicity of biocides: cases from the project COMBASE

Abstract

The evaluation of the ecotoxicological profile of chemicals is of high relevance when a substance can have an impact on the environment, such as the case of biocides. Due to the high number of animal tests conducted each year for regulatory purposes and the ethical considerations that this entails, the requirement of alternative methods by companies and regulatory agencies is increasing. Within these, in silico tools are useful to minimize time, costs, and resources, and they can be applied as alternatives to traditional laboratory assays. In this chapter, we present some computational models developed in the context of the EU LIFE+ project entitled “Computational tool for the assessment and substitution of biocidal active substances of ecotoxicological concern (COMBASE)” (http://www.life-combase.com). The main objective of the project was the development of a tool based on computational toxicology, integrating predictive models of the toxic effects associated with biocidal substances at different trophic levels. Here, different quantitative structure-activity relationship (QSAR) models for the estimation of ecotoxicity of biocides in microorganisms and fish are presented. First, an integrated model to predict the respiratory inhibition in activated sludge was developed, by combining sequentially a qualitative and a quantitative QSAR model. Previously to the development of the model, a set of 94 chemicals with known EC50 values was selected to this study, based on their “biocide-like” structural features. Second, a model to predict LC50 on rainbow trout was developed on a dataset made by collection data from OpenFoodTox database of the European Food Safety Authority (EFSA) and Pesticide Ecotoxicity Database of Office of Pesticide Programs (OPP) (https://ecotox.ipmcenters.org/). Both models showed good performances and robustness and have been integrated in the VEGA last release (version 1.1.5; https://www.vegahub.eu/) as well as the specific COMBASE tool (http://webtool.life-combase.com).