Machine Learning Approach for Determining the Formation of β-Lactam Antibiotic Complexes with Cyclodextrins Using...

Area of Science:

• Pharmaceutical Sciences
• Analytical Chemistry
• Computational Chemistry

Background:

• Cyclodextrins (CDs) are widely used to improve drug properties.
• Understanding the interactions between CDs and β-lactam antibiotics is crucial for drug formulation.
• Nondestructive analytical methods are needed for efficient complex characterization.

Purpose of the Study:

• To develop a machine learning model for detecting β-lactam antibiotic-cyclodextrin complex formation.
• To identify specific molecular interactions involved in complexation.
• To provide a rapid, nondestructive method for complex analysis.

Main Methods:

• Preparation of complexes between various β-lactam antibiotics and cyclodextrins.
• Characterization using differential scanning calorimetry (DSC) for confirmation.
• Analysis using transmission Fourier-transform infrared (tFTIR) and attenuated total reflectance FTIR (ATR) spectroscopy.
• Application of machine learning algorithms for spectral data analysis.
• Molecular modeling using Parameterized Method 7 (PM7) to support experimental findings.

Main Results:

• The machine learning model achieved 90.4% cross-validation accuracy in distinguishing complexed from non-complexed samples.
• Spectroscopic analysis revealed interactions primarily involving the ester groups of β-lactam antibiotics with cyclodextrins.
• Specific interactions were noted with the cephem ring of cefetamet pivoxil and the penam moiety of pivampicillin.
• Molecular modeling provided insights into potential binding modes and explained experimental observations.

Conclusions:

• Machine learning applied to tFTIR and ATR spectra offers a powerful, nondestructive method for determining cyclodextrin complex formation with β-lactam antibiotics.
• The study elucidated specific interaction sites, advancing the understanding of drug-cyclodextrin complexation mechanisms.
• This approach facilitates the development and quality control of pharmaceutical formulations involving cyclodextrin inclusion complexes.