Artificial intelligence-guided biosynthesis and retrosynthesis in pharmacognosy: Toward the synthesis-oriented discovery of natural-product therapeutics

Abstract

Artificial intelligence is reshaping Pharmacognosy by connecting ethnobotanical knowledge, multi-omic data, Biosynthetic pathway prediction, Retrosynthetic planning, and medicinal chemistry optimization. Particular attention is given to AI-driven tools, including BioNavi-NP, graph-sequence-enhanced transformers, NAG2G, RSGPT, RetroExplainer, and human-in-the-loop systems such as DeepRetro. These platforms can reconstruct natural-product biosynthesis, predict plausible precursors, preserve molecular topology, suggest multi-step disconnections, and explore broad reaction spaces. They are especially relevant for metabolites with dense stereochemistry, unusual ring systems, multifunctional scaffolds, and enzyme-guided biosynthetic logic. Beyond route design, AI may help prioritize biosynthetic genes, optimize scarce plant-derived compounds, and guide the development of more drug-like analogues with improved potency, selectivity, pharmacokinetic behavior, and synthetic accessibility. However, important limitations remain, including limited plant-specific reaction datasets, weak reaction-condition prediction, incomplete stereochemical and regioselective modeling, benchmark weaknesses, and the need for expert validation. Overall, AI is best understood as a decision-support layer linking biodiversity, traditional knowledge, biosynthetic logic, and experimental synthesis for responsible future therapeutic discovery and validation across modern natural-product-based drug discovery pipelines.