Abstract: Objective: To explain the changes of composition and medicinal properties of Gastrodia elata during cooking by metabonomics, network pharmacology and molecular docking techniques. Methods: The components of Gastrodia elata were detected by LC-QTOF-MS and analyzed by R language. The bioactive ingredients were screened through PubChem database and literature. Swiss target prediction database and SEA database were used to predict the potential targets of the active ingredients, while GeneCards database and OMIM database were used to search for diseases-related targets. The STRING 11.5 database and Cytoscape3.8.2 software were used to construct drug-compound-target network and target protein interaction network. GO enrichment and KEGG pathway annotation analysis of intersection targets were performed by DAVID database and OmicShare Tools. Results: 89 Components with different contents were found in cooked Gastrodia elata and fresh Gastrodia elata. 11 Compounds were screened for the treatment of convulsion by 410 potential targets, and 5 compounds were screened for the treatment of diabetes and its complications by 698 potential targets. GO and KEGG enrichment analysis showed that Gastrodia elata treated convulsion through IL-17 signaling pathway, cAMP signaling pathway, neuroactive substance receptor signaling pathway, Alzheimer's disease signaling pathway and dopaminergic synaptic signaling pathway, while treated diabetes through AGE-RAGE signaling pathway, HIF-1, VEGF and TNF signaling pathway. The results of molecular docking showed that the selected components were all bound to the cavity of the target protein, and the binding free energy was less than 0 kcal/mol, and the molecular docking conformation was stable. Conclusion: Cooked Gastrodia elata has better neuroprotective effect, and fresh Gastrodia elata has better effect in treating diabetes mellitus and cardiovascular diseases.