基于比值统计法的全基因组差异表达基因识别.docVIP

  基于比值统计法的全基因组差异表达基因 识别# 吉国力1，郑健体1，吴小惠1，李庆顺2* 5 10 （1. 中国厦门大学自动化系，福建 厦门 361005； 2. 美国迈阿密大学植物系，Oxford OH 45056） 摘要：基于高通量和高密度的 tiling array 芯片数据，本文提出了芯片数据预处理方案和基于比 值的差异表达基因识别。首先应用 VSN 标准化算法有效消除了芯片内部探针的背景信号和多芯 片间的系统偏差；再利用 Mummer 软件过滤掉了在拟南芥基因组序列中非唯一完美匹配的探针， 消除了部分不合理探针的干扰；最终提出基于比值的统计法筛选差异表达基因，从全基因组角度 分析和比较拟南芥野生型样本(WT)和病态突变型样本(pcfs4)之间的基因表达差异，筛选出了 24 个统计显著性的差异表达基因，可为生物湿实验提供有价值的参考。 关键词：tiling array；基因表达差异；多聚腺苷化 中图分类号：Q-332 15 Genome-wide recognition of differentially expressed genes by ratio-based statistic method Ji Guoli1, Zheng Jianti1, Wu Xiaohui1, Li Qingshun2 (1. Department of Automation, Xiamen University, FuJian XiaMen 361005; 20 25 30 35 40 2. Department of Botany, Miami University, Oxford, OH 45056, USA) Abstract: Based on the high-throughput and high-density tiling array data, we propossed a chip data preprocessing pipeline and developed a ratio-based statistic method to identify differentially expressed genes. First we applied the the VSN normalization algorithm to effectively eliminate the system deviation between the background signal of the probe of the internal chip and multi-chips. Then we adopted the Mummer software to filter out probes that are not perfectly matched in the Arabidopsis genome sequence. Finally we proposed a statistical method based on the ratio for identification of the differentially expressed genes from the whole genome level to select differentially expressed genes between wild-type (WT) and the mutant (pcfs4). We selected 24 genes with statistically significance, which could provide valuable reference for the biological wet-lab. Keywords: tiling array; differential gene expression; polyadenylation 0 引言 Tiling array( 高通量覆瓦式 基因芯片 )实验技术 是 从传统的微阵列 基因芯片 技术 (Expression Microarray)发展而来，采用无偏的芯片设计思路，从全基因组概念上将寡核苷酸 探针像瓦片一样以固定步长覆盖整个基因组序列。其高密度的探针使得每个转录单位或注释 单位可以被多个探针覆盖，从而可以从统计学的角度估计该单位的表达丰度。如果一个基因 编码多个转录产物，根据不同转录产物上探针信号强度的变化情况还可以揭示相应基因的结 构[1]。正是这些特点使得基于 Arabidopsis tiling 1.0R array 芯片数据，在揭示突变体样本 pcfs4 中表达水平有差异的基因(Differentially expressed