【作者】 孟祥光；

【导师】 周宏灏；

【作者基本信息】 中南大学 ， 临床药理学， 2012， 博士

【摘要】 自从人类基因组计划实施以来,生物信息学得到了迅猛发展。它主要是利用数学、信息学、统计学、计算机等学科来研究生物学的问题,是分子生物学与信息技术(尤其是互联网技术)的结合体。其研究材料是基于计算机的海量生物学数据,其方法就是对生物学数据的收集、筛选、编辑、整理、管理、显示、计算及模拟。‘生物信息学极大的影响了基因组学的一个分支学科——药物基因组学。在它的影响下,药物基因组学以前所未有的速度发展。然而,药物基因组学研究所产生的试验数据大多是分散和无结构化的,所以生物学家们要想获取集中有序的数据异常艰难。如果建立各种便于访问的药物基因组数据库则可以有效的解决这一问题。因此,我们研发了中国人药物基因组信息库(CPGxNet)。此外,我们还进行了一项关于早期肾移植术后病人CYP3A和ABCB1基因多态以及其组成的单倍型对环孢素A药代动力学影响的试验性研究。研究中,我们把CYP3A和ABCB1的分型结果用一种试探性的推测方法顺次和来自Web of Science、CPGxNet数据库中的相关数据进行了比较分析,以此来验证CPGxNet数据库报表数据的可靠性。本课题具体研究结果如下：1、开发了基于浏览器／服务器(B/S)架构完全免费的CPGxNet数据库,并通过互联网进行正式发布(网址：http://www.cpgxnet.net/);2、CPGxNet数据库目前共收录5个基因,98个单核苷酸多态位点(SNP),124种药物,1145条中国人药物相关基因突变记录,并还在持续收录中；3、本数据库中国人样本集信息以及由此的报表功能目前还未发现类似药物基因组数据库涉及,经分析验证可以认为这一模块对研究人员具有一定的参考价值,即可以帮助科研人员弄清楚当前药物基因组学在中国人方面的研究热点,比如涉及的种族、地域、大小、频率、表型、研究时间等；4、我们把试验分型数据(CYP3A4*1G、CYP3A5*3C、ABCB11236C＞T、2677G＞T/A和3435C＞T的突变频率分别为26.2%、69%、65.5%、42.5%/14.3%和38.9%)和来自Web of Science、CPGxNet数据库的相关结果进行比较,验证了CPGxNet数据库生成报表数据的可靠性,提示了CPGxNet数据库具有一定的参考价值；5、我们把对数据库验证的分型数据通过单基因分析,进一步观察到CYP3A5*3C多态性与CYP3A5-CYP3A5AA单倍型和早期肾移植病人服用CsA的药代动力学具有相关性(P<0.05),而未发现CYP3A4*1G、ABCB11236C＞T、ABCB12677G>T/A和ABCB13435C>T多态性以及CYP3A4-CYP3A5GG、ABCB11236-2677-3435TTT、CGC和TGC单倍型和病人的CsA药代动力学有相关性(P>0.05)：6、我们为了确定CYP3A5*3C是否作为独立因素影响药物CsA的药代动力学,我们以CYP3A5*3C、年龄、血红蛋白、血尿素氮和血肌酐为自变量构建的多元线性回归模型,进一步观察到CYP3A5*3C和CsA的药代动力学具有最强的相关性(P<0.05)。更多还原

【Abstract】 The rapid development of bioinformatics profits from the Human Genome Project. The bioinformatics is a kind of combination of molecular biology with information technology, especially on Internet, taking advantage of mathematics, informatics, statistics and computer to solve the biological problems. It is used for dealing with a huge of biological data by collection, processing, computing, simulation and so forth.The pharmacogenomics as a branch discipline of genomics has been greatly affected by bioinformatics so that it is developing at an unprecedented speed. However, the trial data produced by the studies on pharmacogenomics is mostly disperse and unstructured, and thus it is very difficult that biologists attempt to access them. If it is possible to develop all kinds of pharmacogenomics databases accessed easily to, the problem can be effectively solved. Therefore, we established the Chinese Pharmacogenomics Network (CPGxNet).In addition, we evaluate the effect of CYP3A and ABCB1gene polymorphisms on cyclosporine A (CsA) pharmacokinetics in the post-renal transplant recipients by a clinical trial. In the present study, we first compared the results from our clinical trial with the data from Web of Science, and then with the correlated results from the CPGxNet database by a kind of tentative reckoning method to validate the reliability of the CPGxNet report forms data.Our findings are as follows:1. Have developed a completely free CPGxNet database based on the Browse/Server (B/S) structure, and the online site is http://www.cpgxnet.net/. 2. At present, collection information in the CPGxNet included5genes,98single nucleotide polymorphisms (SNP),124kinds of drugs and1145variant records. The update in the database will continue in future.3. The Chinese sample sets and the report form function of the CPGxNet database play an important role in reference for the researchers. The module can help researchers make it clear the current research hot spots on pharmacogenomics in Chinese, which involved in the race, region, size, frequency, phenotype, and research time.4. We compared the results from our trial for126Chinese post-renal transplant recipients with the data including in Web of Science and in the CPGxNet, and determined that the data of the CPGxNet report forms were reliable. Therefore, we draw a conlusion that the CPGxNet database has an important reference value.5. By the single gene analysis, we observed the correlation between CYP3A5*3C/CYP3A4-CYP3A5AA and CsA pharmacokinetics in early post-renal transplant recipients (P＜0.05), and but we did not find that the correlation between CYP3A4*1G, ABCB11236C＞T,2677G>T/A,3435C＞T/CYP3A4-CYP3A5AA, ABCB11236-2677-3435and CsA pharmacokinetics (P>0.05);6. By constructing a multivariate linear regression model, we found that there still existed this correlation between CYP3A5*3C and CsA pharmacokinetics (P＜0.05).更多还原