A hybrid machine learning and molecular dynamic simulation approaches for identifying FtsZ inhibitors in Staphylococcus aureus

The development of antibiotics with novel mechanisms of action is required due to the significant hazard to public health posed by the global increase in multidrug-resistant (MDR) Staphylococcus aureus infections. FtsZ, a tubulin-like GTPase that is essential for bacterial cytokinesis and is absent in human cells, is a highly selective antibacterial target. A comprehensive in silico methodology was implemented in this study to identify natural product-based inhibitors that target the FtsZ protein in S. aureus . Lipinski's Rule of Five was employed to curate and refine a library of natural compounds in order to identify drug-like candidates. These were subjected to molecular docking, which yielded 1070 compounds with strong binding affinities (