PREMIER partners publish new scientific paper: The VERA software: Implementation of the acute fish toxicity endpoint and its application to pharmaceutical compounds 

Read-across is a widely employed non-testing method (NAM) to assess a target molecule in a data gap situation. It provides (eco)toxicological information about the target substance using the property values of similar compounds, considering mainly structural features as a possible measurement of similarity.  

One of the aims of the PREMIER project is to apply this methodology to assess the acute fish toxicity of various Active Pharmaceutical Ingredients (APIs). This type of assessment is highly challenging considering the limited ecotoxicological information available for pharmaceuticals. Indeed, experimental data primarily exist for more generic substances such as pesticides, biocides, and industrial products. Furthermore, APIs generally have a more complex chemical structure with consequent more elaborate biological mechanisms. Finally, these approaches have the limitation of not being easily reproducible, making the assessment of a target substance sensitive to the expert who has applied such methodology. 

For this reason, we have developed a new approach for the read across, the Virtual Extensive Read-Across (VERA), which provides the assessment of a target substance in an automated and thus reproducible manner. It utilizes various similarity metrics simultaneously to enhance the search for similar compounds, even for more complex substances such as APIs. Specifically, it employs a global similarity score, molecular groups, and structural alerts to identify clusters of similar substances; these clusters are then used to identify suitable analogues and achieve the assessment for the target API. 

See the article here.