| Lung cancer,as one of the most common malignant tumors in the world today,seriously affects human life health,among which non-small cell lung cancer accounts for the vast majority of lung cancer.Compared with western medicine for the treatment of lung cancer,the use of traditional Chinese medicine for the treatment of lung cancer has many advantages,such as improved patient survival quality,reduced cancer recurrence rate,and prolonged patient survival.However,TCM pays attention to the dialectical thinking of the holistic view,but it is still not sufficiently clear at the level of the mechanism of action and molecular mechanism in treating diseases,while virtual screening technology,based on drug design theory with the aid of computer computing power and professional application software,can pick out some slim compounds from a large number of compounds and is able to analyze the relationship between TCM’s component compounds and disease targets from a microscopic perspective,Interpretation and discovery of drugs in the treatment of non-small cell lung cancer.However,TCM ingredients are complex,and their pharmacological effects often exhibit the characteristics of multi-target and multi-action,so that the application of virtual screening technology in the field of TCM has different characteristics from western medicine.Therefore,in this paper,we conducted a relevant research and analysis of TCMs for the treatment of NSCLC by combining bioinformatics and virtual screening techniques to find possible relevant genes and relevant anti NSCLC TCMs.And to analyze and summarize the applications of virtual technology among the fields of traditional Chinese medicine,to analyze the problems and applications of virtual screening among the fields of traditional Chinese medicine.Moreover,the prediction problems from ingredients to TCM are discussed and the related algorithms are improved,aiming to provide a reference for the application of virtual screening in TCM.The main work of this paper is as follows:Firstly,Using bioinformatics methods,several gene sets related to non-small cell lung cancer were obtained from a comprehensive gene expression database,and these data were mined and processed to screen out potential genes related to non-small cell lung cancer.After verification,10 key genes,including KIF20 A,NUF2,TTK,TPX2,TOP2 A,CDC6,DLGAP5,NCAPG,CCNB1,KIF11,were identified that may be highly related to non-small cell lung cancer,Exploring the relationship between these genes and non-small cell lung cancer.Secondly: through the literature analysis method,the literatures of several virtual screening technologies and TCM applications from 2000 to 2021 were collected,the applications of virtual screening technologies in TCM fields were analyzed and summarized,the applications of virtual screening technologies in TCM were divided according to different technologies and different fields,and the relevant involved databases and data platform sources were counted,Finally,the problems and characteristics of virtual screening applied in the field of traditional Chinese medicine are analyzed.Thirdly: To explore and analyze the problem from TCM ingredients to TCM prediction,firstly,to analyze the current status and deficiency of existing algorithms,and to explore the optimization side focus of problems and algorithms that the TCM prediction problem should pay attention to,and to propose improvements and Optimization for related algorithms with respect to possible directions of improvement,an improved algorithm based on Manhattan distance combining the best distance proportion and weight of active ingredients is proposed.And design a contrast experiment is constructed to put forward the performance metric indexes of the algorithm according to the optimization side of the algorithm,and the alignment of the algorithm according to the metric indexes is performed,and it is found that the improved algorithm is more suitable for TCM prediction problems.Fourthly: using virtual screening technology,from different databases to obtain the required data and materials,and to optimize the appropriate screening model,the ingredients of traditional Chinese medicine highly related to the relevant targets of non-small cell lung cancer are screened out.And based on these ingredients,we used an improved algorithm to screen out traditional Chinese medicine(TCM)against NSCLC,and obtained 10 herbs,such as semen secundii,Fructus evodiae,Radix ephedra rhizome,Croton cream,radix rehmanniae,Radix magnoliae,Radix magnoliae,root of Polygonum macrophyllae,Radix ephedra officinale,and Dendrobium Dendrobii,which were associated with three genotypes of TTK,TOP2 A and KIF11. |