微阵列数据揭示基因在癌样本中广泛表达

摘要

The gene expression data are subject to multiple sources of technical artifacts due to experimental differences in sample preparation, array processing and others. Hence, data normalization is supposed to adjust the global properties of measurements of individual samples so that they can be "more appropriately compared". Based on the assumption that only a few genes are differentially expressed in a disease and have balanced upward and down- ward expression level changes, researchers usually normalize microarray data by forcing all of the arrays to have the same probe intensity distributions to remove technical variations in the data. However, accumulated evidences suggest gene expressions could be widely altered in cancer, so we need to evaluate the sensitivities of biological discoveries to violation of the normalization assumption. In this study, by analyzing two expression profiles of pancreatic cancer, we showed that the distribution of gene expression in cancer samples are significantly different from that in normal samples, indicating that the assumption of current normalization is unreliable for pancreatic cancer. Our results showed that using normalized data may falsely produce many down - regulated differential expressed genes while missing many up - regulated differential expressed genes in pancreatic cancer. Furthermore, using the raw intensities for analysis, we found extensive up - regulation of gene expression in pancreatic cancer, which provieds important hints for studying the mechnism of pancreatic cancer.%癌的发生与发展过程涉及大量基因的异常表达.在目前基因表达谱分析中采用的标准化方法通常假设在疾病中差异表达的基因的比例很小并且差异上、下调的比例大致相等.这个被研究者所广泛采用的标准化的前提假设尚未被充分地论证过.通过分析胰腺癌的两套表达谱数据,我们发现在胰腺癌样本中基因表达的中值显著高于正常样本,提示传统的标准化假设并不适用于胰腺癌表达谱数据.采用标准化数据会导致错误地判断大量的差异下调的基因并失查许多差异上调的基因.采用原始数据分析发现在胰腺癌中的基因表达有广泛上调的特征,为深入研究胰腺癌的发生和发展机制提供了新线索.