Emerging Role of Artificial Intelligence in Natural Product-Based Drug Discovery
Background: Natural products remain an important source of therapeutic agents, but their discovery is constrained by chemical complexity, time-consuming isolation, difficult structure elucidation, and high attrition rates. Artificial intelligence (AI), including machine learning, deep learning, graph neural networks, and generative AI, offers computational approaches to address these challenges. Method: This narrative review examines peer-reviewed, English-language literature published between 2021 and 2026, retrieved from PubMed, Scopus, ScienceDirect, and Google Scholar. The selected literature was analyzed using thematic qualitative synthesis to identify AI applications, reported contributions, and persistent limitations across the natural product discovery pipeline. Problem Identification: Despite growing interest in AI-assisted drug discovery, the extent to which current methods address the limitations of conventional approaches remains uncertain. Key concerns include data quality, model reliability, interpretability, and the need for experimental validation. Results: AI applications cover target identification, biosynthetic gene cluster mining, virtual screening, structure elucidation, network pharmacology, generative molecular design, and absorption, distribution, metabolism, excretion, and toxicity (ADMET) prediction. These methods can help prioritize candidate compounds and predict potential biological activities. However, data scarcity and bias, inconsistent evaluation practices, limited interpretability, and insufficient experimental validation remain significant barriers. Explainable AI, multi-omics integration, and self-driving laboratories represent promising developments, although their practical contributions require further validation. Conclusion: AI can support and accelerate selected stages of natural product-based drug discovery but cannot replace experimental testing or the expertise of natural product chemists and pharmacologists.
Authors
- Yelfi Anwar (ORCID: https://orcid.org/0000-0002-6161-4781)
Institutions
- Universitas 17 Agustus 1945 Jakarta (ID)
Publication Details
- Journal
- Medical Jurnal Kesehatan dan Kedokteran
- Published
- 2026-11-04
- DOI
- https://doi.org/10.69836/medical-jkk.v3i2.985
- Primary Topic
- Computational Drug Discovery Methods
- Type
- article
- Field-Weighted Citation Impact
- 0.00