A team of researchers at the University of Vienna have developed an innovative software tool, MolCompass, to identify the weaknesses of machine learning models used for risk assessment of chemical compounds. This tool aims to increase transparency and confidence in these models, which have become increasingly popular in recent years. The results of this research have been published in the Journal of Cheminformatics. The University of Vienna is also a member of the RISK-HUNT3R project, which aims to develop non-animal risk assessment methods for new chemicals.
