Traditional Medicine Knowledge Graphs for Drug Discovery: Cultural Context, Evidence Grading, Mechanistic Links, and Benefit-Sharing Logic

Download PDF

Traditional medicine knowledge has long informed natural-product research, yet its representation in computational drug discovery raises scientific, cultural, and governance challenges. Knowledge graphs may help organize heterogeneous information concerning traditional uses, botanical identity, preparation context, phytochemical composition, bioactivity, molecular targets, pathways, safety, and literature provenance. However, graph construction can also detach knowledge from its cultural setting, merge incompatible evidence types, expose sensitive information, overstate mechanistic certainty, or support extractive research practices. This article proposes a responsible knowledge framework for traditional medicine knowledge graphs in drug discovery. The framework places cultural context preservation, source provenance, sensitive knowledge triage, evidence grading, citation traceability, uncertainty annotation, and conflicting-evidence management before mechanistic interpretation or candidate prioritization. It further distinguishes reported traditional use, published ethnopharmacological evidence, experimental bioactivity, computational inference, mechanistic hypotheses, safety evidence, and validated therapeutic claims. Benefit-sharing logic, access governance, ethical sourcing, community-interest recognition, expert and stakeholder review, and feedback updating are incorporated as explicit framework dimensions rather than treated as external considerations. The proposed architecture is intended to support respectful, transparent, and evidence-aware hypothesis generation. It does not establish therapeutic efficacy, confirm biological mechanism, certify ethical or legal compliance, represent community approval, or demonstrate commercialization readiness. Its principal contribution is an integrated conceptual structure for connecting traditional medicine knowledge with modern drug-discovery evidence while preserving epistemic boundaries, cultural meaning, governance responsibilities, and validation requirements.