| ObjectiveBased on the clinical data of insomnia,from the perspective of biomolecular interaction network between disease,symptom clusters phenotype and herb target,this paper explored the classification method of symptom clusters,the mining method of effective prescriptions and the biomolecular network association mechanism of symptom clusters.By this means,we attempted to interpret the scientific connotation of "disease,syndrome,treatment and effect",so as to provide methodologies for related research.Methods(1)Methods to judge the effective case of insomnia clinical data.We used the structured data and adopted the unified evaluation standard and quality control method,to analyze the characteristics and efficacy of insomnia patients in different cohort.At the same time,the propensity score matching method was used to eliminate the confounding factors such as age,course of disease,education degree and so on,so as to obtain a balanced and effective sample subgroup.(2)Methods for classification of insomnia symptom clusters and mining effective prescriptions.The text mining method was used to process symptoms from effective insomniacs to extract high-quality standardized symptom set;Then,latent structure analysis was carried out by using Lantern app to explore the classification and identification rules of insomnia symptom clusters.Gephi and Liquorice app were used to mine the effective core prescription,and OR and X2 test were used to analyze the correspondence between the effective prescription and the symptom clusters.(3)Network pharmacology analysis method was used.First,we refered to UMLS to standardize related terms.Then,the insomnia gene,symptom gene and herb compound target in biomedical database were integrated.Finally,the PPI network was constructed on STRING platform.And the visual display and network topology analysis of PPI network were carried out by using Cytoscape app,and the KEGG and Gene Ontology enrichment analysis of hub node were carried out on David platform.It was necessary to preliminarily verify the results,and the methods included:?insomnia symptom gene,herb target and drug target take intersection;?herb target mapping the pathway of insomnia drug;?the intersection of symptom gene network and herb target network's hub node enriches KEGG pathway.(4)Methods of analyzing differential genes of insomnia symptom clusters.RNA-seq was used to analyze the gene differential expression between insomnia and healthy people.The KEGG and Gene Ontology enrichment analysis was carried out on cluster Profiler to explore whether the biomolecular network association mechanism of different symptom clusters,and the results of RNA-seq and data mining were used to verify each other.Results(1)Analysis of characteristics and evaluation of effect of insomnia patients in different cohort:Baseline characteristics of insomnia patients were analyzed,such as gender,accompanying anxiety and depression,etc.And the results showed that there were 994 patients and 2697 visit in 9 cohort.There were gender differences,and female were more than male.In addition,most of them were brainwork.They all had high education level,and most of them lived together.Anxiety and depression were common emotional problems,and their daytime functions were impaired.According to the improvement of PSQI and SE,the efficacy evaluation criteria were established.And the results showed that the highest effective rate was 91.9%in physician ?,followed by 63.5%in physician H,62.4%in physician F and 62.1%in physician D.A total of 500 effective cases and 471 ineffective cases were matched with propensity score of 1:1,and 340 effective and 340 ineffective subgroups were obtained.(2)Classification of insomnia symptom clusters and discovery of effective prescriptions:Five common symptom clusters and effective prescriptions were obtained.?Symptom cluster I was composed of dizziness,anorexia,belching,palpitation,fatigue and so on,which reflected the pathogenesis of deficiency of both heart and spleen,and was treated with Astragalus,Angelica sinensis,Paeonia lactiflora Pall,Lilium lancifolium Thunb,Ziziphi Spinosae Semen,etc.?Symptom cluster ? was composed of fatigue,irritability,headache,deep-colored urine and so on,which reflected the pathogenesis of liver depression transforming into fire,and was treated with Cyperus rotundus L,Gardenia jasminoides Ellis,Curcuma wenyujin,Scutellaria baicalensis Georgi,Nacre,etc.?Symptom cluster ? was composed of dizziness,soreness and weakness of waist and knees,dry mouth,tinnitus and so on,which reflected the pathogenesis of disharmony of heart and kidney,and was treated with Lilium lancifolium Thunb,Anemarrhena asphodeloides Bge,Caulis Polygoni Multiflori,Ziziphi Spinosae Semen,etc.?Symptom cluster ? was composed of tongue ecchymosis,anorexia,nightmares and so on,which reflected the pathogenesis of blood stasis,and was treated with Bupleurum chinense DC,Rehmannia glutinosa Libosch,Angelica sinensis,Paeonia veitchii,Cyathula officinalis Kuan,ect.?Symptom cluster V was composed of bitter taste in mouth,dizziness,oppression in chest,phlegm,deep-colored urine and so on,which reflected the pathogenesis of phlegm heat,and treated with Coptis chinensis Franch,Lapis Chloriti,Arum Ternatum Thunb,Bamboo Shavings,Aurantii Fructus Immaturus,ect.(3)Research on the discovery of biomolecular network associated with insomnia symptom clusters and effective prescription:The biomedical database was integrated to construct the biomolecular network of Symptom cluster ? which reflected the pathogenesis of deficiency of both heart and spleen,which contained 320 genes,while the formula contained 60 compounds and 932 targets.Preliminary validation showed that there were 33 identical targets and 95 identical pathways for symptom clusters and the formula,9 identical targets for the formula and insomnia drugs,and the formula mainly acted on GABAergic synapses,phenylalanine metabolism pathway,etc.While there were 430 genes in the biomolecular network of Symptom cluster ? which reflected the pathogenesis of liver depression transforming into fire,and 124 compounds and 1221 targets were included in the formula.The preliminary verification showed that there were 32 identical targets and 108 identical pathways between the symptom clusters and the formula,16 identical targets between the prescription and insomnia drugs,and the prescription mainly acted on serotonergic synapse,linoleic acid metabolism and so on.(4)RNA-seq analysis of small sample and healthy control of insomnia symptom cluster:A total of 22 healthy controls,21 patients of Symptom cluster ? which reflected the pathogenesis of deficiency of both heart and spleen,and 11 patients of Symptom cluster ? which reflected the pathogenesis of liver depression transforming into fire were analyzed by RNA-seq.The results showed that 67 genes were unique in patients of Symptom cluster I,and their GO terms were mainly related to water and electrolyte transmembrane transport,skeletal muscle thin filament assembly,etc.Besides,their KEGG pathways were mainly related to Circadian entrainment,digestion and absorption of nutrients,and regulation of saliva and insulin secretion.However,there were 149 unique differential genes in patients of Symptom cluster ?.In addition,their GO terms were mainly related to cell differentiation,NOS activity regulation,and KEGG pathways were mainly related to calcium signaling pathway,circadian rhythm,Serotonergic synapse,etc.(5)Research on double verification between RNA-seq and data mining:For one thing,the differential gene set and data mining gene set of patients of Symptom cluster I which reflected the pathogenesis of deficiency of both heart and spleen contained 7 identical genes.And the relevant formula included 11 herbs,which contained 17 compounds could map 9 differential gene,such as astragaloside ?,jujuboside A,ferulic acid,paeoniflorin,etc.The hub node of data mining gene set and differential gene set contained 72 identical KEGG pathways.On the other hand,the differential gene set and data mining gene set of patients of Symptom cluster II which reflected the pathogenesis of liver depression transforming into fire contained 8 identical genes.And the relevant formula included 13 herbs,which contained 43 compounds could map 28 differential gene,such as Baicalein,geniposide,cyperolone,etc.The hub node of data mining gene set and differential gene set contained 54 identical KEGG pathways.ConclusionThis study broke through the traditional research ideas of molecular biological mechanism of syndrome,drew lessons from the thought of "correspondence of prescription and syndrome"in Treatise on febrile diseases.And the international concept of symptom clusters was introduced,which was relatively fixed and highly generalized,and might objectively reflected syndrome.In addition,the network pharmacology and RNA-seq were used to corroborate each other,and we established a set of methods to explore the mechanisms underlying the molecular network associations between symptom clusters and corresponding formula.Generally,we hope for providing methodologies for the research on elucidating the scientific connotation of"disease,syndrome,treatment and effect".(1)Based on the real-world clinical data of insomnia,the effective cases were screened according to the accepted curative effect criteria,and then the confounding factors were eliminated by propensity score matching,which provided methodologies for analyzing and mining the characteristics of effective prescriptions and symptom clusters.(2)It showed that the key point of the study was the classification of symptom clusters,and the latent structure analysis might suitable for the recognition and classification of insomnia symptom clusters.And it might provide a mature model and algorithm for the objectivity and stability of symptom clusters.(3)In this study,network pharmacology was used to establish the relationship between TCM symptom clusters and gene,which broke through the bottleneck of the lack of corresponding relationship.Combined with the RNA-seq,it confirmed that the protocol was basically feasible to explore the biomolecular network association of insomnia symptom clusters and corresponding effective prescription.(4)This study initially found that different symptom clusters of insomnia had different molecular network association mechanism,and it might relate to the physiology and pathology of viscera and be able to explain the scientific connotation of pathogenesis.It showed that the exploration of syndrome differentiation and treatment based on relatively fixed symptom clusters and its molecular network association mechanism combined with "pathogenesis" was the further deepening of the research idea of "disease,syndrome,treatment and effect". |
PDF Full Text Request