Abstract. Discovering drugs to a disease is still a challenging task for researchers due to the complexity of biomolecules involved in pathologic processes. Design and development of new and more efficient drugs is still urgent for several diseases. Cheminformatics tools are useful to better understand the complex structures of chemical compounds and the implication of chemical features in the activity. In the current work, a series of synthetic 3-arylcoumarins, with reported antioxidant activity, was studied. A virtual screening, based on the TOPSMODE approach, using a clastogenic model, was performed to predict the potential genotoxicity of the studied molecules. A preliminary interpretation of the relationship between structure and clastogenicity suggests the importance of hydroxyl groups at positions 7 and/or 8 of the coumarin ring. This communication is focused on cheminformatics and its applications on drug discovery, helping to find solutions to complex diseases.