In silico prediction of receptor-mediated environmental toxic phenomena-application to endocrine disruption.

It is an objective of our institution to establish a virtual laboratory allowing for a reliable in silico estimation of the harmful effects triggered by drugs, chemicals and their metabolites. In the recent past, we have developed the underlying technology (Multi-dimensional QSAR: Quasar and Raptor) and compiled a pilot system including the 3D models of three receptors known to mediate endocrine-disrupting effects (the aryl hydrocarbon receptor, the estrogen receptor and the androgen receptor, respectively) and validated them against 310 compounds (drugs, chemicals, toxins). Within this set up we could demonstrate that our concepts are able to both recognize toxic compounds substantially different from those used in the training set as well as to classify harmless compounds clearly as being non-toxic. This suggests that our approach can be used for the prediction of adverse effects of drug molecules and chemicals.