QSAR and machine learning applied for the analysis of (fluoro)quinolone activity

Fluoroquinolones (FQs) are a class of antibiotics effective against both Gram-positive and Gram-negative bacteria owing to their ability to target DNA gyrase and topoisomerase IV. The growth of bacterial resistance to antimicrobials leads to greater attention to comparative evaluation of biological activities of pharmaceutical preparations. In this regard, a quantitative structure-activity relationship (QSAR) analysis is used for predicting FQ activity. QSAR serves as an effective tool for predicting antibacterial, antiviral, anti-cancer, genotoxic activity, etc. Moreover, the QSAR approach provides possibilities to estimate FQ contribution to the environment and ecosystems.

This review summarizes more than 100 publications on QSAR and machine learning studies of various medicinal, chemical, and biological characteristics of FQs.

The authors expect the appearance of novel FQs with improved bactericidal activity. This will be done with in silico assistance of computer-aided drug design and artificial intelligence techniques in drug development. Future QSAR models will be effectively applied to three aspects of FQ activity: 1) detection of FQs with the use of immunoassays; 2) high photo- and biodegradability of antibiotics in the environment; 3) physicochemical activity, in particular photochemical production of singlet oxygen and free radical species.