Wu Gang, Nie Lei, Freeland Stephen J
Department of Biological Sciences, University of Maryland at Baltimore County, Baltimore, MD 21250, USA.
Biochem Biophys Res Commun. 2007 Jul 13;358(4):1108-13. doi: 10.1016/j.bbrc.2007.05.043. Epub 2007 May 15.
It is well-established that non-random patterns in coding DNA sequence (CDS) features can be partially explained by translational selection. Recent extensions of microarray and proteomic expression data have stimulated many genome-wide investigations of the relationships between gene expression and various CDS features. However, only modest correlations have been found. Here we introduced the one-way ANOVA, a more powerful extension of previous grouping methods, to re-examine these relationships at the whole genome scale for Saccharomyces cerevisiae, where genome-wide protein abundance has been recently quantified. Our results clarify that coding sequence features are inappropriate for use as genome-wide estimators for protein expression levels. This analysis also demonstrates that one-way ANOVA is a powerful and simple method to explore the influence of gene expression on CDS features.
众所周知，编码DNA序列（CDS）特征中的非随机模式可以部分地由翻译选择来解释。微阵列和蛋白质组表达数据的最新扩展激发了许多关于基因表达与各种CDS特征之间关系的全基因组研究。然而，只发现了适度的相关性。在这里，我们引入了单向方差分析，这是对先前分组方法更强大的扩展，以在全基因组规模上重新审视酿酒酵母中这些关系，最近已对其全基因组蛋白质丰度进行了量化。我们的结果表明，编码序列特征不适用于作为全基因组蛋白质表达水平的估计指标。该分析还表明，单向方差分析是一种强大而简单的方法，可用于探索基因表达对CDS特征的影响。
