Discovery of Xuantoujiedu Decoction and its Molecular Mechanisms Using Integrated Network Analysis

Traditional Chinese medicine (TCM) has accumulated a lot of practical experience in the treatment of respiratory infections, which incorporates many effective clinical formulae for specific disease conditions. Here, we distilled the real-world clinical data derived from a highly experienced TCM physician, Professor Xue Boshou, who has particular empirical knowledge for treating respiratory infections. Using the 9584 clinical prescritions from Xue, we constructed a prescription network (PN) with nodes representing precriptions and links representing the similarity of prescriptions. After detecting the prescription modules of the PN, we found that one significant prescription cluster (containing 1020 prescripitions) was to treat respiratory infections (with main indications for fever and sore throat). Furthermore, we obtained 277 medical cases with fever and sore throat and detected an effective clinical formula, called Xuantoujiedu Decoction(XD), with 13 core herb ingredients. In addition, By using Weighted Point Mutual Information, we identified the herb modification patterns between symptoms and herbs. The herb composition and indications of XD were clearly defined by incorporating the empirical knowledge of Xue. By integrating the 312 drug targets of XD and symptom gene assocations, we further elucidated the underlying molecular mechanisms of XD with significant pharmacological insights. Overall, we demonstrated an integrated network analysis approach to find effective clinical formula from real-world clinical data and further elucidate the pharmacological mechanisms of the detected formula.