Development of clinical genome information systems for pharmacokinetics analysis

我们利用OGSA-WebDB技术开发了一个临床基因组信息网格检索系统，该系统可以从两个或多个生命科学数据库（db）中提取全面的药物相关信息。检索DB包括TOXINET（药物不良反应文件信息数据库）、PharmGKB（药物相关基因数据库）、Entrez Protein（蛋白质信息数据库）、LSBM（基因表达器官数据库）、KEGG（途径信息数据库）和JSNP（单核苷酸多态性信息数据库）。该系统可以提取最新的药物责任基因和器官、代谢途径数据以及与抗癌或其他药物不良反应相关的蛋白质信息，研究人员可以估计药物责任基因与药物不良反应之间的因果关系。以此验证盐酸伊立替康与顺铂联合治疗的药物不良反应。常见的途径是通过细胞色素P450参与外源代谢的ABC转运体，淀粉和蔗糖代谢，并提取了56个snp作为相关多态性。基于临床病例资料，建立盐酸伊立替康的药代动力学分析算法。我们开发了临床基因组药代动力学模拟系统，在选择基因型（野生型、杂合子、纯合子）、身高、体重和实际给药剂量后，可以将CPT-11、SN-38和SN-38G在各室的浓度随时间变化以图表的形式可视化。它还可以模拟^*28和^*6的UGT1A1等位基因型模式。我们提出了临床基因组信息管理系统的系统规范，该系统包括临床基因组数据、患者病例数据和药代动力学数据。原型系统是在UNIX服务器上使用Apache2.2.0、PHP5.1.1、MySQL5.0.18和Java编程语言开发的。

We developed a clinical genome information grid search system that can extract comprehensive drug-related information from two or more life-science databases (DBs) using OGSA-WebDB technology, which is a database grid technology. The retrieval DBs are TOXINET (an adverse-drug-reaction document information DB), PharmGKB (a drug-related gene DB), Entrez Protein (a protein information DB), LSBM (a gene-expression organ DB), KEGG (a pathway-information DB), and JSNP (a single nucleotide polymorphism (SNP) information DB). The system can extract references to the newest drug responsibility genes and organs, metabolic pathway data, and protein information relevant to adverse reactions to anti-cancer or other drugs, and researchers can estimate the causal relationships between a drug responsibility gene and an adverse drug reaction. Using this, we verified the adverse drug reaction between irinotecan hydrochloride and cisplatin in combination therapy. The common pathways are the ABC transporters that are involved in the metabolism of xenobiotics by cytochrome P450, and starch and sucrose metabolism and 56 SNPs were extracted as relevant polymorphisms. A pharmacokinetics analysis algorithm for irinotecan hydrochloride was created based on clinical case data. We developed a clinical genome pharmacokinetics simulation system in which the chronological concentration change of CPT-11, SN-38, and SN-38G in each compartment can be visualized as a graph after selecting the genotype (wild-type, heterozygote, homozygote), body height, weight, and actual drug dose. It can also simulate the UGT1A1 allele type patterns of ^*28 and ^*6. We propose the system specifications for a clinical genome information-management system that includes clinical genome data, patient case data, and pharmacokinetics data. The prototype system was developed using Apache2.2.0, PHP5.1.1, MySQL5.0.18, and Java programming language on a UNIX server.