Artificial Intelligence-Assisted In Silico Drug Discovery of Bioactive Compounds from SpathodeaCampanulataAgainst Antimicrobial Resistance

Abstract

Antimicrobial resistance (AMR) is one of the most serious global health challenges, threatening the effectiveness of existing antibiotics and increasing mortality rates worldwide. The discovery of novel antimicrobial agents from medicinal plants offers a promising strategy to combat resistant pathogens. Spathodeacampanulata (African Tulip Tree) is a medicinal plant known for its antimicrobial, antioxidant, and anti-inflammatory properties. This study explores the potential of phytochemicals derived from Spathodeacampanulata as inhibitors of antimicrobial resistance-associated targets through artificial intelligence (AI)-assisted in silico drug design. A computational workflow involving phytochemical screening, machine learning-based activity prediction, molecular docking, drug-likeness assessment, and ADMET profiling was employed. Several compounds including quercetin, luteolin, kaempferol, apigenin, and ursolic acid demonstrated promising interactions with bacterial targets such as DNA gyrase, β-lactamase, dihydrofolatereductase, and penicillin-binding protein 2a. The AI model identified compounds with high antimicrobial probability scores, while docking studies indicated strong binding affinities. ADMET evaluation suggested favorable pharmacokinetic properties for selected molecules. These findings support the potential application of S. campanulata phytochemicals as lead candidates for the development of new antimicrobial agents against resistant bacterial strains.