基于随机森林算法的赤芍成分-靶点-心脑血管疾病网络药理作用研究
|
摘要
目的:采用随机森林算法,构建中药赤芍成分-靶点-疾病网络,预测赤芍治疗心脑血管类疾病的关键靶标。方法:搜集KEGG数据库中小分子药物及其酶、离子通道、G蛋白、核蛋白等4类药靶数据作为训练集,建立基于随机森林法的药物-靶点相互作用模型,采用十折交叉验证评价模型精度;利用该模型对已报道的赤芍化学成分进行靶点预测,并构建其成分-靶点-疾病网络。结果:4类药靶模型的预测正确率分别为71. 36%,67. 08%,73. 71%,67. 83%;从赤芍的9个化学成分预测出多个作用靶点,预测结果得到了较好的文献验证。结论:所建模型具有较高的预测精度。
Objective: To predict key targets of Paeoniae Radix Rubra on cardio-cerebral vascular diseases and to construct the corresponding compound-target-disease network of Paeoniae Radix Rubra. Methods: Small molecule drugs and the related enzymes,ion channels,G-protein-coupled receptors,nuclear receptors downloaded from KEGG database were used as the training set. The drug-target interaction model was built by random forest algorithm,and the accuracies were evaluated by 10-fold cross-validation tests. The model was then applied to predict the potential targets interacted with the reported compounds of Paeoniae Radix Rubra and to construct a compound-target-diseases network. Results: The prediction accuracies of the four models were 71. 36%,67. 08%,73. 71% and 67. 83%,respectively. Multiple targets were predicted from the 9 chemical components of Paeoniae Radix Rubra,The predicted results were well verified by literatures. Conclusion: The models established achieve high prediction accuracies.
Objective: To predict key targets of Paeoniae Radix Rubra on cardio-cerebral vascular diseases and to construct the corresponding compound-target-disease network of Paeoniae Radix Rubra. Methods: Small molecule drugs and the related enzymes,ion channels,G-protein-coupled receptors,nuclear receptors downloaded from KEGG database were used as the training set. The drug-target interaction model was built by random forest algorithm,and the accuracies were evaluated by 10-fold cross-validation tests. The model was then applied to predict the potential targets interacted with the reported compounds of Paeoniae Radix Rubra and to construct a compound-target-diseases network. Results: The prediction accuracies of the four models were 71. 36%,67. 08%,73. 71% and 67. 83%,respectively. Multiple targets were predicted from the 9 chemical components of Paeoniae Radix Rubra,The predicted results were well verified by literatures. Conclusion: The models established achieve high prediction accuracies.
引文
[1]国家药典委员会.中华人民共和国药典[S]. 2015年版.一部.北京:中国医药科技出版社,2015:158.
[2]陆小华,马骁,王建,等.赤芍的化学成分和药理作用研究进展[J].中草药,2015,46(4):595-602.
[3]王博龙,刘志强.计算机模拟研究冠心宁注射液主要成分治疗心血管疾病的网络药理学机制[J].中草药,2018,49(14):3357-3368.
[4]王奕森,夏树涛.集成学习之随机森林算法综述[J].信息通信技术,2018,12(1):49-55.
[5]赵芳卿,翟菲,项荣武,等.网络药理学研究态势的可视化分析[J].中华中医药杂志,2018,33(7):3099-3103.
[6] KANEHISA M,GOTO S,HATTORI M,et al. From genomics to chemical genomics:new developments in KEGG[J]. Nucleic Acids Res,2006,34(Database issue):D354-D357.
[7]魏玉辉.化学计量学在中药复杂体系研究中的应用[D].兰州:兰州大学化学化工学院,2010:13-14.
[8]刘艳丽.随机森林综述[D].天津:南开大学数学科学学院,2008:5-11.
[9]郑丽丽,阎西艴,张永珍,等.赤芍对高脂兔血小板胞浆游离钙、红细胞膜Ca2+,Mg2+-ATP酶活性的影响[J].中国中西医结合杂志,1996,16(5):295-296.
[10]管谷爱子.芍药的药理及药效[J].现代东洋医学,1991,12(4):93-102.
[11] MAO QQ,IP SP,TSAI SH,et al. Antidepressant-like effect of peony glycosides in mice[J]. J Ethnopharmacol,2008,119(2):272-275.
[12]金英善,陈曼丽,陶俊.芍药化学成分和药理作用研究进展[J].中国药理学与毒理学杂志,2013,27(4):745-750.
[13] GUO RB,WANG GF,ZHAO AP,et al. Paeoniflorin protects against ischemia-induced brain damages in rats via inhibiting MAPKs/NF-кB-mediated inflammatory responses[J]. PLoS One,2012,7(11):e49701.
[14]饶梦琳,唐蜜,何锦悦,等.芍药苷对大鼠局灶性脑缺血再灌注脑血流量及PGI_2/TXA_2平衡的影响[J].药学学报,2014,49(1):55-60.
[15]杨军,王静,姜文,等.赤芍总苷对D_半乳糖衰老小鼠学习记忆及代谢产物的影响[J].中国药理学通报,2001,17(6):697-699.
[16]吴芳.赤芍总苷/淫羊藿总黄酮对实验性抑郁及脑5-HT和β-肾上腺素受体的影响[J].现代预防医学,2005,32(7):744-746.
[2]陆小华,马骁,王建,等.赤芍的化学成分和药理作用研究进展[J].中草药,2015,46(4):595-602.
[3]王博龙,刘志强.计算机模拟研究冠心宁注射液主要成分治疗心血管疾病的网络药理学机制[J].中草药,2018,49(14):3357-3368.
[4]王奕森,夏树涛.集成学习之随机森林算法综述[J].信息通信技术,2018,12(1):49-55.
[5]赵芳卿,翟菲,项荣武,等.网络药理学研究态势的可视化分析[J].中华中医药杂志,2018,33(7):3099-3103.
[6] KANEHISA M,GOTO S,HATTORI M,et al. From genomics to chemical genomics:new developments in KEGG[J]. Nucleic Acids Res,2006,34(Database issue):D354-D357.
[7]魏玉辉.化学计量学在中药复杂体系研究中的应用[D].兰州:兰州大学化学化工学院,2010:13-14.
[8]刘艳丽.随机森林综述[D].天津:南开大学数学科学学院,2008:5-11.
[9]郑丽丽,阎西艴,张永珍,等.赤芍对高脂兔血小板胞浆游离钙、红细胞膜Ca2+,Mg2+-ATP酶活性的影响[J].中国中西医结合杂志,1996,16(5):295-296.
[10]管谷爱子.芍药的药理及药效[J].现代东洋医学,1991,12(4):93-102.
[11] MAO QQ,IP SP,TSAI SH,et al. Antidepressant-like effect of peony glycosides in mice[J]. J Ethnopharmacol,2008,119(2):272-275.
[12]金英善,陈曼丽,陶俊.芍药化学成分和药理作用研究进展[J].中国药理学与毒理学杂志,2013,27(4):745-750.
[13] GUO RB,WANG GF,ZHAO AP,et al. Paeoniflorin protects against ischemia-induced brain damages in rats via inhibiting MAPKs/NF-кB-mediated inflammatory responses[J]. PLoS One,2012,7(11):e49701.
[14]饶梦琳,唐蜜,何锦悦,等.芍药苷对大鼠局灶性脑缺血再灌注脑血流量及PGI_2/TXA_2平衡的影响[J].药学学报,2014,49(1):55-60.
[15]杨军,王静,姜文,等.赤芍总苷对D_半乳糖衰老小鼠学习记忆及代谢产物的影响[J].中国药理学通报,2001,17(6):697-699.
[16]吴芳.赤芍总苷/淫羊藿总黄酮对实验性抑郁及脑5-HT和β-肾上腺素受体的影响[J].现代预防医学,2005,32(7):744-746.