Quinolones are characterized by a broad spectrum of antimicrobial activity, favorable pharmacokinetic properties and low toxicity. The potential increase in resistance to quinolones, forces the manufacturers to carry out work on new drug substances. In a few works there have been attempts to find a correlation between the characteristics of the chemical structure and the value of minimal inhibitory concentration (MIC50) of quinolones. Purpose of this study is to determine the relationship between physicochemical parameters of quinolones and the MIC50 values ​​ designated for Haemophilus spp., Moraxella spp., Neisseria spp., Proteus vulgaris, Serratia spp., Shigella spp., Staphylococcus spp., Streptococcus spp., Yersinia spp. Analysis of physicochemical parameters of selected drugs was made using MarvinSketch 5.11.5 (ChemAxon Ltd.) and QuickProp 3.1 software from Schrödinger package v 31207. MIC50 values were correlated with of the 51 physicochemical parameters calculated. The leave-one-out (LOO) method was used for model cross-validation. The calculations were made in relation to the average value MIC50 7 strains of bacteria Gram – and 2 strains of bacteria Gram +. A validation was carried out between proposed arithmetic expressions in relations to average values of MIC50 calculated only for these bacteria. It was shown that the analysis of the structure: MIC50 correlation, with regard to many bacterial strains requires the binding of many physicochemical structure parameters in the form of arithmetic expressions. Only a combination of physicochemical structure characteristics in the form of arithmetic expressions allows reflecting the complex interactions between the bacterial cell and drug molecules.