REVISION OF THE DUAL SCALING METHOD FOR SUCCESSIVE CATEGORIES DATA AND ITS APPLICATION TO QUANTITATIVE STRUCTURE-ACTIVITY-RELATIONSHIPS (QSAR)

Nishisato's Dual Scaling Method for successive categories data has been revised in order to apply it to quantitative structure-activity relationships. The revised version can give category values of successive categories, even though the original one can give only boundary values between two adjacent successive categories. The new version can be applied to an analysis of the antimicrobial activities of polymethylene bi(pyridineamine) derivatives, which are potential agents for controlling and/or preventing the formation of dental plaque. The results show that the proper molecular length of the agents leads to the maximum antimicrobial activity.