# DENDRAL **Domain:** Artificial Intelligence / Chemistry / Scientific Discovery **Doc Type:** Canonical System Node **Maturity:** Foundational ## Definition **DENDRAL** was a Stanford research program that used chemical constraints and mass-spectrometry data to generate and evaluate candidate molecular structures. It is widely recognized as the first expert system for scientific hypothesis formation. ## Architecture DENDRAL's importance lies in the relationship between search and knowledge. A combinatorial generator could propose structures, but encoded chemical rules made the search tractable by excluding implausible candidates and interpreting evidence. The project joined Joshua Lederberg, Edward Feigenbaum, Bruce Buchanan, Carl Djerassi and other collaborators. Its early motivation connected automated chemical analysis with space biology and remote life-detection questions. ## Evolutionary Nexus [[articles/The Evolutionary Roots of Silicon Valley|The Evolutionary Roots of Silicon Valley]] treats DENDRAL as a founding convergence of biology and practical AI. The system did not merely borrow biological language; it grew from a geneticist's scientific problem and a chemist's domain knowledge. ## Continuity Context DENDRAL demonstrates reconstruction without identity: molecular structure can be inferred from fragments, but the resulting hypothesis is a model of the source molecule, not its continuation. The same distinction governs reconstructed persons. ## Sources / Provenance - Stanford Biomedical Data Science history: https://dbds.stanford.edu/biomedical-data-science-graduate-program-overview/ - Buchanan et al., “DENDRAL: A Case Study of the First Expert System for Scientific Hypothesis Formation,” publication record: https://cs.stanford.edu/people/eaf/wordpress/cv/ ## See Also [[wiki/Expert System|Expert System]], [[wiki/Reasoning Engine|Reasoning Engine]], [[wiki/Encoded Domain Knowledge|Encoded Domain Knowledge]], [[wiki/Exobiology|Exobiology]], [[wiki/MYCIN|MYCIN]]