Consideration of RNA Secondary Structure Significantly Improves Likelihood-Based Estimates of Phylogeny: Examples from the Bilateria

Abstract

Sequences from ribosomal RNA (rRNA) genes have made a huge contribution to our current understanding of metazoan phylogeny and indeed the phylogeny of all of life. That said, some parts of this rRNA-based phylogeny remain unresolved. One approach to increase the resolution of these trees would be to use more appropriate models of sequence evolution in phylogenetic analysis. RNAs transcribed from rRNA genes have a complex secondary structure mediated by base pairing between sometimes distant regions of the rRNA molecule. The pairing between the stem nucleotides has important consequences for their evolution which differs from that of unpaired loop nucleotides. These differences in evolution should ideally be accounted for when using rRNA sequences for phylogeny estimation. We use a novel permutation approach to demonstrate the significant superiority of models of sequence evolution that allow stem and loop regions to evolve according to separate models and, in common with previous studies, we show that 16-state models that take base pairing of stems into account are significantly better than simpler, 4-state, single-nucleotide models. One of these 16-state models has been applied to the phylogeny of the Bilateria using small subunit rRNA (SSU) sequences. Our optimal tree largely echoes previous results based on SSU in particular supporting the tripartite Bilaterian tree of deuterostomes, lophotrochozoans, and ecdysozoans. There are also a number of differences, however, perhaps most important of which is the observation of a clade consisting of the gastrotrichs plus platyheminthes that is basal to all other lophotrochozoan taxa. Use of 16-state models also appears to reduce the Bayesian support given to certain biologically improbable groups found using standard 4-state models.

Variation patterns of the mitochondrial 16S rRNA gene with secondary structure constraints and their application to phylogeny of cyprinine fishes (Teleostei: Cypriniformes)

“…Second, the effect of selective pressures on parsimony analysis is loose (Van de Peer, 1993;Sloss et al, 2004). Finally, the down-weighting schemes of non-independent characters are too simplified and do not elucidate the different selective pressures in stem regions (Telford et al, 2005), and thus are not adequate for real data sets. Therefore, our results suggest that the down-weighting of characters of stems relative to loops might not be helpful to improve phylogenetic inference.…”

“…Some recent analyses using the mitochondrial 16S rRNA gene, however, have depicted limited resolution of relationships within the Cyprinidae because of low nodal support (Gilles et al, 1998;Simons and Mayden, 1998;Simons et al, 2003;Li et al, 2005). It is well known that phylogenetic inferences depend not only on the appropriate choice of the marker, but also rely on comparative studies of the inherent phylogenetic contents of the molecular markers, such as base compositions, structural characteristics, parameter estimations and the adopted methods (Rosenberg and Kumar, 2003;Telford et al, 2005). One approach to improve the phylogenetic performance of the mitochondrial 16S rRNA gene is to incorporate information regarding the molecular structure of the marker in analyses for more accurate phylogenetic inference.…”

Geographical patterns of deep mitochondrial differentiation in widespread Malagasy reptiles

“…Separate evolution models for stem and loop regions of rRNA genes are preferable over a single evolution model for the entire sequence (Telford et al, 2005). We therefore divided sequences in stem and loop partitions based on published secondary structure models.…”

Phylogeny of sipunculan worms: A combined analysis of four gene regions and morphology

“…The stem regions were analyzed under a doublet model with a single rate parameter and sixteen states (Schöninger and von Haeseler, 1994), representing all possible nucleotide pairings. Currently, the doublet model cannot be tested for goodness of Wt against other models with common software programs, such as Modeltest (Posada and Crandall, 1998) or MrModeltest (Nylander, 2004), but recent studies have shown that the 16-state doublet model signiWcantly outperformed simpler, 4-state, single-nucleotide models for stem regions of ribosomal sequences (Savill et al, 2001;Telford et al, 2005). No secondary structure information was used for the fragment of 28S rRNA because (1) no appropriate sipunculan reference sequence was available, and (2) only ca.…”

“…Better resolution and support for deep relationships was achieved by Telford et al (2005) who presented a phylogeny of the Bilateria, based on 72 complete 18S ribosomal RNA sequences. The sequences were analyzed under mixed models for the stem and loop regions of the ribosomal molecule, applying a doublet model that accounts for correlation of substitutions in the complementary stem regions.…”