Rough sets in the analysis of the structure-activity relationships of antifungal imidazolium compounds

The relationship between chemical structure and antifungal activity of quaternary imidazolium compounds is analyzed. The compounds are described by a set of condition attributes concerning structure and by a decision attribute concerning activity. The description builds up an information system. The smallest set of condition attributes, significant for high-quality classifications, has been found by using the rough sets approach. The analysis of the distribution of the value of the significant condition attributes in the best and the worst classes lead to the definition of typical representatives of the best and the worst imidazolium chlorides in terms of the significant condition attributes. A decision algorithm has been developed from the information system, presenting an important relationship between structure and antifungal activity. This may be helpful in supporting decisions concerning synthesis of new antifungal imidazolium compounds.