Analysis of relationship between whole-genome transcriptome in human tissues and characteristics of the human genome

我们已经获得了人类22个组织、神经干细胞和细胞系的全基因组(4000万个点；1 GB)的表达谱。我们还构建了一个带有注释的全基因组数据的转录组数据库的可视化工具，并分析了人类基因组中转录组的特征。由于与标准微阵列相比，平铺阵列实验需要更高质量的RNA，我们首次开发了一套RNA纯化方法和质量控制系统，以确保所有样品的RNA都是高质量的。此外，我们还开发/修改了全长RNA检测和克隆的方法，并通过qRT-PCR证实了其表达。对于表达谱，我们提出了一种新的、广泛适用的归一化方法，通过优化参数，使其在基因表达…实验数据的基础上，获得最低的假阴性率和最高的灵敏度更多的离子微阵列。利用这些，我们新发现人类基因组中的大多数区域在每个组织中都有表达，并且超过90%的转录组来自非编码区。我们还鉴定了数以万计的Transcipt产物，它们在每个组织中特异表达。通过对人类基因组注释数据的分析，我们发现了人类基因组中表达模式、序列保守性和多态变异之间的有趣关系。这些为聚焦区域进行功能分析提供了新的方法。这个人类组织全基因组表达数据的数据库在世界上是独一无二的。研究人员将能够通过将人类组织中的表达数据与疾病细胞(例如癌细胞)的表达数据进行比较，找到包括非编码区在内的疾病相关基因。特别是，这个数据库对于搜索组织特定的疾病相关分子很有用，也可以通过观察正常组织中的表达来避免副作用，从而优先选择治疗的候选分子。较少

We have obtained whole-genome (40 million points; 1Gb) expression profiles of 22 tissues, neural stem cells, and cell-lines of the human. We also constructed a visualization tool for the transcriptome database acompanied with annotated genome-wide data, and analyzed characteristics of the transcriptome in the human genome. Because a higher quality set of RNA was required for the tiling array experiments compared to standard microarray ones, we originally developed an RNA purification methodology and a quality control system, with which we conformed that the RNA from all the samples were with high quality. Also, we developed/modified protocols for full-length RNA detection and cloning, and confirmed the expressions by qRT-PCR. For the expression profiling, we developed a new and widely applicable normalization method by optimizing parameters so that they would achieve the least false negative rate and the highest sensitivity on the basis of additional experimental data from gene-express … More ion microarrays. Using these, we newly found that most regions in the human genome were expressed in each tissue and that more than, 90% of transcriptome were from non-coding regions. We also identified dozens of thousands of transcipt products that expressed specifically in each tissue. By analysis with human genome annotation data, we found interesting relationship between the expression patterns, sequence conservation, and variation of polymorphisms in the human genome. These provide new methodologies for focusing regions for functional analyses. This database of genome-wide expression data in the human tissues is unique in the world. Researchers will be able to find disease-related loci including non-coding regions by comparing the expression data in human tissues with those from disease cells, e.g. cancer cells. Particularly, this database is useful for searching tissue specific disease-related molecules and also for prioritizing candidate molecules for therapies by looking at expression in normal tissues to avoid side effects. Less