基于数据挖掘和生物信息学探讨专利中药复方治疗酒精性肝病的用药规律及作用机制

Medication Rule and Mechanism of Patented Chinese Herbal Compounds in Treating Alcoholic Liver Disease Based on Data Mining and Bioinformatics

专利中药复方是实用性较强的验方,但其治疗酒精性肝病(alcoholic liver disease,ALD)的用药规律和作用机制尚不明确。采用数据挖掘方法探讨中药专利复方治疗ALD的用药规律,运用生物信息学探索核心药对调节铁死亡治疗ALD的潜在作用机制。通过中医传承计算平台(V3.0)对治疗ALD的专利中药复方进行数据挖掘,提取核心药对。从GSE28619数据集、FerrDb和BATMAN-TCM数据库中获取“核心药对-ALD-铁死亡”共同分子靶点的差异表达基因(differentially expressed genes,DEGs)并分析其相关性和富集分析。构建机器学习模型筛选关键DEGs,构建列线图模型预测ALD的患病风险并验证。共获得229个专利中药复方,包含609味中药。高频药物有丹参、甘草等23味,以补虚药、利水渗湿药、清热药和活血化瘀药为主,丹参-山楂是核心药对。共有10个共同分子靶点DEGs,它们具有相互调控、介导促性腺激素释放激素、糖尿病并发症中晚期糖基化终产物及其受体等信号通路的作用。RF机器学习模型的预测效果最佳,磷酸丝氨酸转氨酶1(phosphoserine aminotransferase 1,PSAT1)、早期生长反应因子1(early growth response factor 1,EGR1)、醌氧化还原酶1(NADPH quinone oxidoreductase 1,NQO1)、醛酮还原酶1C3(aldo-keto reductase 1C3,AKR1C3)、JUN原癌基因(Jun proto-oncogene,JUN)是关键的DEGs。列线图模型能够准确地预测ALD的患病风险。结果表明:中药专利复方治疗ALD以扶正祛邪为基本原则,用药以补虚药为主,兼以利水渗湿、清热、活血化瘀等药物。核心药对丹参-山楂调节铁死亡治疗ALD可能与PSAT1、EGR1、NQO1等DEGs及其介导的促性腺激素释放激素、糖尿病并发症中晚期糖基化终产物及其受体等信号通路有关。

Patent Chinese herbal compounds are empirical prescription with strong practicability,but its medication rules and mechanism in the treatment of alcoholic liver disease(ALD)are still unclear.To analyze the medication regularity of patented Chinese herbal compounds in ALD and explore the potential mechanism of core drugs in regulating ferroptosis in treating ALD by using the methods of data mining and bioinformatics.Using Traditional Chinese Medicine Inheritance Computing System(V3.0),data mining was carried out on the Chinese herbal compounds for the treatment of ALD,and the core drug pairs were extracted.The differentially expressed genes(DEGs)of common molecular targets of“core drug pairs-ALD-ferroptosis”were obtained from GSE28619 dataset,FerrDb and BATMAN-TCM database.The correlation and enrichment of DEGs were analyzed.The machine learning model was constructed to screen the key genes of DEGs,and the nomogram model to predict the incidence risk of ALD was constructed and verified.A total of 229 Chinese herbal compounds were obtained,including 609 traditional Chinese medicines.There were 23 kinds of high-frequency drugs,such as Salviae Miltiorrhizae Radix et Rhizoma,Glycyrrhizae Radix,which are mainly tonifying deficiency drugs,promoting diuresis and dampness drugs,clearing heat drugs and promoting blood circulation and removing blood stasis drugs.Salviae Miltiorrhizae Radix et Rhizoma-Hawthorn is the core drug pair in the treatment of ALD.A total of 10 common molecular targets DEGs were obtained,which can regulate each other and mediate GnRH,AGE-RAGE and other signaling pathways.The RF machine learning model has the best prediction effect,and PSAT1,EGR1,NQO1,AKR1C3 and JUN are the key DEGs.The nomogram model can accurately predict the risk of ALD.The results show that the basic principle of treating ALD with patented Chinese herbal compounds is to strengthen the vital energy and eliminate pathogenic factors.In terms of medication,tonifying deficiency drugs are the main drugs,and promoting diuresis and dampness,clearing heat,promoting blood circulation and removing blood stasis drugs are the auxiliary drugs.Salviae Miltiorrhizae Radix et Rhizoma-Hawthorn regulating ferroptosis in the treatment of ALD may be related to the DEGs such as PSAT1,EGR1,NQO1 and their mediated GnRH,AGE-RAGE and other signaling pathways.