In the chemical and pharmaceutical industries, many procedures are based on the principle that compounds with similar structures tend to exhibit similar activities. Therefore, the development of computational methods for decision support in similarity assessment is of utmost importance. In this proof-of-concept study, we introduce a computational framework for quantitative computation of agreement between structural and physiological similarity of compounds. As model molecules, we use bridged heterometallic complexes. To derive physiological binary vectors we use rate constants for the first substitution reaction between our compounds and some biomolecules. It is believed that such treatment of metal complexes allows a better understanding of the structure-activity relationship and can serve as a guide for new synthetic targets.