Jin Ping, Zhao Yingdong, Ngalame Yvonne, Panelli Monica C, Nagorsen Dirk, Monsurró Vladia, Smith Kina, Hu Nan, Su Hua, Taylor Phil R, Marincola Francesco M, Wang Ena
Immunogenetics Section, Department of Transfusion Medicine, Clinical Center, NIH Bethesda, MD 20892, USA.
BMC Genomics. 2004 Aug 13;5(1):55. doi: 10.1186/1471-2164-5-55.
Endogenous reference genes are commonly used to normalize expression levels of other genes with the assumption that the expression of the former is constant in different tissues and in different physiopathological conditions. Whether this assumption is correct it is, however, still matter of debate. In this study, we searched for stably expressed genes in 384 cDNA array hybridization experiments encompassing different tissues and cell lines.
Several genes were identified whose expression was highly stable across all samples studied. The usefulness of 8 genes among them was tested by normalizing the relative gene expression against test genes whose expression pattern was known. The range of accuracy of individual endogenous reference genes was wide whereas consistent information could be obtained when information pooled from different endogenous reference genes was used.
This study suggests that even when the most stably expressed genes in array experiments are used as endogenous reference, significant variation in test gene expression estimates may occur and the best normalization is achieved when data from several endogenous reference genes are pooled together to minimize minimal but significant variation among samples. We are presently optimizing strategies for the preparation of endogenous reference gene mixtures that could yield information comparable to that of data pooled from individual endogenous reference gene normalizations.
内源性参照基因常用于标准化其他基因的表达水平，其假设是前者在不同组织和不同生理病理条件下表达恒定。然而，这一假设是否正确仍存在争议。在本研究中，我们在涵盖不同组织和细胞系的384次cDNA阵列杂交实验中寻找稳定表达的基因。
鉴定出了几个在所研究的所有样本中表达高度稳定的基因。通过针对表达模式已知的测试基因对相对基因表达进行标准化，对其中8个基因的实用性进行了测试。单个内源性参照基因的准确范围很广，而当使用来自不同内源性参照基因的汇总信息时，可以获得一致的信息。
本研究表明，即使在阵列实验中使用最稳定表达的基因作为内源性参照，测试基因表达估计值仍可能出现显著差异，当将来自多个内源性参照基因的数据汇总在一起以最小化样本间最小但显著的差异时，可实现最佳标准化。我们目前正在优化制备内源性参照基因混合物的策略，该混合物可产生与来自单个内源性参照基因标准化的数据相当的信息。
