Unleashing the Power of Artificial Intelligence-Driven Drug Discovery in Streptomyces

Abstract

The rise of antibiotic resistance has created an urgent need for the discovery of new antibiotic compounds. Streptomycin, the first antibiotic isolated from Streptomyces sp., paved the way for discovering other antibiotics for combating bacterial infections. By exploring the genome-based biosynthetic potential of various Streptomyces species, a vast array of secondary metabolites with potential therapeutic applications can be identified, contributing a transformative impact on the field of medicine. However, conventional screening approaches on novel natural products (NPs) from Streptomyces sp. have entered a bottleneck due to inefficiency. Fortunately, artificial intelligence (AI) and machine learning (ML) models enable rapid exploration and prediction of potential antibiotic compounds, increasing the probability of discovering new antibacterial compounds. AI-driven drug discovery in Streptomyces sp. represents a paradigm shift in the future quest for novel pharmaceutical agents. Various ML models have been developed and applied in different practical applications. Overall, the ML model is trained using input data and generates outcomes based on prediction output. This review discusses the continued potential of Streptomyces sp. as a source of novel NPs, along with the application of ML throughout the NP drug discovery pipeline involving genome mining, biological activities prediction, and optimization compound production in Streptomyces microbial systems. Graphical abstract: The role of machine learning in drug discovery from Streptomyces