BackgroundWithin the complex metazoan phylogeny, the relationships of the three lophophorate lineages, ectoprocts, brachiopods and phoronids, are particularly elusive. To shed further light on this issue, we present phylogenomic analyses of 196 genes from 58 bilaterian taxa, paying particular attention to the influence of compositional heterogeneity.ResultsThe phylogenetic analyses strongly support the monophyly of Lophophorata and a sister-group relationship between Ectoprocta and Phoronida. Our results contrast previous findings based on rDNA sequences and phylogenomic datasets which supported monophyletic Polyzoa (= Bryozoa sensu lato) including Ectoprocta, Entoprocta and Cycliophora, Brachiozoa including Brachiopoda and Phoronida as well as Kryptrochozoa including Brachiopoda, Phoronida and Nemertea, thus rendering Lophophorata polyphyletic. Our attempts to identify the causes for the conflicting results revealed that Polyzoa, Brachiozoa and Kryptrochozoa are supported by character subsets with deviating amino acid compositions, whereas there is no indication for compositional heterogeneity in the character subsets supporting the monophyly of Lophophorata.ConclusionOur results indicate that the support for Polyzoa, Brachiozoa and Kryptrochozoa gathered so far is likely an artifact caused by compositional bias. The monophyly of Lophophorata implies that the horseshoe-shaped mesosomal lophophore, the tentacular feeding apparatus of ectoprocts, phoronids and brachiopods is, indeed, a synapomorphy of the lophophorate lineages. The same may apply to radial cleavage. However, among phoronids also spiral cleavage is known. This suggests that the cleavage pattern is highly plastic and has changed several times within lophophorates. The sister group relationship of ectoprocts and phoronids is in accordance with the interpretation of the eversion of a ventral invagination at the beginning of metamorphosis as a common derived feature of these taxa.

Compositional heterogeneity of sequences between taxa may cause systematic error in phylogenetic inference. The potential influence of such bias might be mitigated by strategies to reduce compositional heterogeneity in the data set or by phylogeny reconstruction methods that account for compositional heterogeneity. We adopted several of these strategies to analyze a large ribosomal protein data set representing all major metazoan taxa. Posterior predictive tests revealed that there is compositional bias in this data set. Only a few taxa with strongly deviating amino acid composition had to be excluded to reduce this bias. Thus, this is a good solution, if these taxa are not central to the phylogenetic question at hand. Deleting individual proteins from the data matrix may be an appropriate method, if compositional heterogeneity among taxa is concentrated in a few proteins. However, half of the ribosomal proteins had to be excluded to reduce the compositional heterogeneity to a degree that the CAT model was no longer significantly violated. Recoding of amino acids into groups is another alternative but causes a loss of information and may result in badly resolved trees as demonstrated by the present data set. Bayesian inference with the CAT-BP model directly accounts for compositional heterogeneity between lineages by introducing breakpoints along the branches of the phylogeny at which the amino acid composition is allowed to change but is computationally expensive. Finally, a neighbor joining tree based on equal input distances that consider pattern and rate heterogeneity showed several unusual groupings, which are most likely artifacts, probably caused by the loss of information resulting from the transformation of the sequence data into distances. As long as no more efficient phylogenetic inference methods are available that can directly account for compositional heterogeneity in large data sets, using methods for reducing compositional heterogeneity in the data in combination with methods that assume a stationary amino acid composition remains an option for controlling systematic errors in tree reconstruction that result from compositional bias. Our analyses indicated that the paraphyly of Deuterostomia in some analyses is the result of systematic errors that also affected the relationships of Entoprocta and Ectoprocta.

No abstract

No abstract

BackgroundFor phylogenetic reconstructions, conflict in signal is a potential problem for tree reconstruction. For instance, molecular data from different cellular components, such as the mitochondrion and nucleus, may be inconsistent with each other. Mammalian studies provide one such case of conflict where mitochondrial data, which display compositional biases, support the Marsupionta hypothesis, but nuclear data confirm the Theria hypothesis. Most observations of compositional biases in tree reconstruction have focused on lineages with different composition than the majority of the lineages under analysis. However in some situations, the position of taxa that lack compositional bias may be influenced rather than the position of taxa that possess compositional bias. This situation is due to apparent symplesiomorphic characters and known as "the symplesiomorphy trap".ResultsHerein, we report an example of the sympleisomorphy trap and how to detect it. Worms within Terebelliformia (sensu Rouse & Pleijel 2001) are mainly tube-dwelling annelids comprising five 'families': Alvinellidae, Ampharetidae, Terebellidae, Trichobranchidae and Pectinariidae. Using mitochondrial genomic data, as well as data from the nuclear 18S, 28S rDNA and elongation factor-1α genes, we revealed incongruence between mitochondrial and nuclear data regarding the placement of Trichobranchidae. Mitochondrial data favored a sister relationship between Terebellidae and Trichobranchidae, but nuclear data placed Trichobranchidae as sister to an Ampharetidae/Alvinellidae clade. Both positions have been proposed based on morphological data.ConclusionsOur investigation revealed that mitochondrial data of Ampharetidae and Alvinellidae exhibited strong compositional biases. However, these biases resulted in a misplacement of Trichobranchidae, rather than Alvinellidae and Ampharetidae. Herein, we document that Trichobranchidae was apparently caught in the symplesiomorphy trap suggesting that in certain situations even homologies can be misleading.