Preliminary Research On Network Pharmacology Of Traditional Chinese Medicine For Treating Cardio-cerebral Vascular Diseases

Pharmacological effects of cardio-cerebral vascular medicine have been extensivelyreported. However, most of them are still at the stage of the observation of clinical casesand the determination of the relevant biomarkers and systematic studies in the molecularlevel remain unexplored. Thus, the related pharmacological mechanisms of them needfurther studies. Aiming at this goal, in this study we managed to construct threecomplicated compound-target-cardio-cerebral-vascular-disease networks of Ligusticumchuanxiong, Ginkgo biloba and Erigeron breviscap us, then to predict effectivecomponents and potential targets of Ligusticum chuanxiong, Ginkgo biloba and Erigeronbreviscap us on cardio-cerebral vascular diseases by network pharmacology. We believethat the current study will be of great help for understanding the mechanism of cardio-cerebral vascular Traditional Chinese medicine at the molecular level class and providingmethodology for the modernization of Traditional Chinese medicine.In this study four drug-target interaction models were built by the random forestalgorithm using the drug compounds and the related enzymes, ion channels, G-protein-coupled receptors, nuclear receptors downloaded from KEGG database as the training sets.Three typical categories of cardio-cerebral vascular Traditional Chinese medicineLigusticum chuanxiong, Ginkgo biloba and Erigeron breviscap us as examples, throughchemistry database and PubChem database collecting information about their chemicalcompositions, random forest algorithm was applied to predict the potential targetsinteracted with compounds of cardiovascular medicine, three complicated compound-target-disease networks of cardio-cerebral vascular Traditional Chinese medicine then wereconstructed and analyzed.The models accuracies were evaluated by10-fold cross-validation tests, and theoverall success rates of the four models were71.34%,67.08%,73.17%and67.83%,respectively. Using the present models the predicted targets of26compounds ofLigusticum chuanxiong, the predicted targets of34compounds of G. biloba and thepredicted targets of10compounds of Erigeron breviscap us were validated by literatures. Furthermore, three complicated compound-target-disease networks were then constructedand analyzed., preliminarily revealing the active ingredients and potential targets ofLigusticum chuanxiong, Ginkgo biloba and Erigeron breviscap us in treatment of cardio-cerebral vascular diseases.The models established here achieved high prediction accuracies, and can be utilizedto find potential targets for compounds from other Chinese medicines. The presentapproach may play an important role in studying the material base and the molecularmechanism of traditional Chinese medicine.