29] Modeling RNA tertiary structure from patterns of sequence variation

Abstract

Publisher Summary Comparative sequence analysis (CSA) is the most reliable way to obtain biologically pertinent secondary structures of RNA molecules in the absence of crystallographic data. In addition to the tRNA cloverleaf, the secondary structure models of ribosomal RNAs, self-splicing introns, the RNase P RNA, and many other natural or in vitro selected RNA molecules were all established by CSA. Phylogenetic analyses can be used to infer tertiary interactions as well. Tertiary interactions identified by CSA have been incorporated into current 3-D models of 16S ribosomal RNA and RNase P RNA and even formed the main basis of a model of the three-dimensional architecture of group I self-splicing introns, which has been widely used, and is now known to have been accurate as far as architecture is concerned. CSA and the subsequent modeling of RNA tertiary structure rely on human judgment. CSA can be divided into the following steps. Within a set of related RNA sequences, segments of similar sequence are tentatively assumed to be homologous and aligned; variable sites within or next to these conserved blocks are searched for evidence of concerted base changes; and the statistical constraints, thus, uncovered are then interpreted in terms of potential secondary and tertiary base pairings.