基于探针杂交机制的生物芯片数据预处理统计方法研究

项目名称： 基于探针杂交机制的生物芯片数据预处理统计方法研究

项目编号： No.10871009

项目类型： 面上项目

立项/批准年度： 2009

项目学科： 生物科学

项目作者： 邓明华

作者单位： 北京大学

项目金额： 24万元

中文摘要： 本项目研究跨实验室、跨平台芯片数据预处理与整合问题。旨在发展一套基于芯片探针杂交物理模型的统计方法，考虑到芯片中探针强度对探针序列的相关性以及PCR 扩增所带来的系统偏差，利用单芯片样本估计靶序列拷贝数，实现跨实验室、跨平台芯片数据标准化与整合。作为应用，拟将算法应用于多种芯片数据分析之中，分析包括基因表达芯片、SNP 芯片、Tiling Array 芯片以及外显子芯片(Exon Array)等流行的生物芯片数据，开发多种芯片数据预处理开源软件，为生物学家提供分析工具。

中文关键词： 基因芯片; PDNN 模型; 芯片探针; 靶序列拷贝数

英文摘要： This proposal is focusing on pre-processing and integrated analysis of gene chip data form different laboratory as well as different platform. It's aim at developing statistical methods based on the physical model of probe hybridization, which will consider the dependence of probe intensity with the probe sequences, as well as the systematic bias caused by the PCR amplification. Such a model can estimate the copy number of target sequences from a single chip, so as to achive the inter-laboratory and inter-platform data pre-processing and integrated analysis. As an application, we propose to apply the model to the data anlysis of different gene chips, such as gene expression array, SNP array, tiling array as well as exon array. We also propose to develop the chip data pre-processing software, which will be served as public data analysis tools.

英文关键词： Gene Chip; PDNN model; probe set; Copy number of target sequence