Abstract
Background: Two main factors, which have an influence on oral absorption from solid, immediate release dosage form, are solubility and permeability. These are considered the main fundamental properties that govern the rate and extent of oral absorption. The significance of these properties has been highlighted in the Biopharmaceutics Classification System (BCS).
Objective: The concept of this paper was to predict the solubility and permeability of fluoroquinolones using in silico methods based on the assumptions of the BCS. An attempt was also made to determine the place within this system for drugs from the fluoroquinolone group. Method: The study was carried out with the use of modern computational techniques which developed based on Artificial Neural Network Ensembles for Binary Classification. Results: Using the values of the physicochemical descriptors of medicinal compounds with labeled BCS class, two classification models were elaborated for solubility and permeability. Conclusion: The obtained models helped to predict the provisional class for the following drugs in the BCS. Continuous improvement of computational models may support and can be treated equally with the in vivo data.Keywords: Artificial neural networks, biopharmaceutics classification system, computational models, fluoroquinolones, molecular modeling, structure-properties relationship.
Graphical Abstract
Article Metrics
42
1