Smit Sandra, Yarus Michael, Knight Rob
Department of Chemistry and Biochemistry, Campus Box 215, University of Colorado at Boulder, Boulder, CO 80309, USA.
RNA. 2006 Jan;12(1):1-14. doi: 10.1261/rna.2183806.
We have encountered an unexpected property of rRNA secondary structures that may generalize to all RNAs. Analysis of 8892 ribosomal RNA sequences and structures from a wide range of species revealed unexpected universal compositional trends. First, different categories of rRNA secondary structure (stems, loops, bulges, and junctions) have distinct, characteristic base compositions. Second, the observed patterns of variation are similar among sequences from large and small rRNA subunits and all domains of life, despite extensive evolutionary divergence. Surprisingly, these differences do not seem to be related to selection for different compositions in different structural categories, but rather relate to the overall composition of the molecule: Randomized RNAs with no evolutionary history show the same structure-dependent compositional biases as rRNAs. These compositional trends may improve the accuracy of RNA secondary structure prediction, because they allow us to compare predicted structures against known compositional preferences. They also suggest caution in interpreting differences in the rate of change of the GC content in different parts of the molecule as evidence of differential selection.
我们发现了核糖体RNA二级结构的一个意外特性，该特性可能适用于所有RNA。对来自广泛物种的8892个核糖体RNA序列和结构的分析揭示了意外的普遍组成趋势。首先，不同类别的rRNA二级结构（茎、环、凸起和接头）具有独特的特征性碱基组成。其次，尽管进化差异很大，但在来自大、小rRNA亚基以及生命所有域的序列中，观察到的变异模式相似。令人惊讶的是，这些差异似乎与不同结构类别中不同组成的选择无关，而是与分子的整体组成有关：没有进化历史的随机RNA显示出与rRNA相同的结构依赖性组成偏差。这些组成趋势可能会提高RNA二级结构预测的准确性，因为它们使我们能够将预测结构与已知的组成偏好进行比较。它们还表明，在将分子不同部分中GC含量变化速率的差异解释为差异选择的证据时应谨慎。
