Skip to main page content
Access keys NCBI Homepage MyNCBI Homepage Main Content Main Navigation
, 16, 987

Extracting Phylogenetic Signal and Accounting for Bias in Whole-Genome Data Sets Supports the Ctenophora as Sister to Remaining Metazoa


Extracting Phylogenetic Signal and Accounting for Bias in Whole-Genome Data Sets Supports the Ctenophora as Sister to Remaining Metazoa

Marek L Borowiec et al. BMC Genomics.


Background: Understanding the phylogenetic relationships among major lineages of multicellular animals (the Metazoa) is a prerequisite for studying the evolution of complex traits such as nervous systems, muscle tissue, or sensory organs. Transcriptome-based phylogenies have dramatically improved our understanding of metazoan relationships in recent years, although several important questions remain. The branching order near the base of the tree, in particular the placement of the poriferan (sponges, phylum Porifera) and ctenophore (comb jellies, phylum Ctenophora) lineages is one outstanding issue. Recent analyses have suggested that the comb jellies are sister to all remaining metazoan phyla including sponges. This finding is surprising because it suggests that neurons and other complex traits, present in ctenophores and eumetazoans but absent in sponges or placozoans, either evolved twice in Metazoa or were independently, secondarily lost in the lineages leading to sponges and placozoans.

Results: To address the question of basal metazoan relationships we assembled a novel dataset comprised of 1080 orthologous loci derived from 36 publicly available genomes representing major lineages of animals. From this large dataset we procured an optimized set of partitions with high phylogenetic signal for resolving metazoan relationships. This optimized data set is amenable to the most appropriate and computationally intensive analyses using site-heterogeneous models of sequence evolution. We also employed several strategies to examine the potential for long-branch attraction to bias our inferences. Our analyses strongly support the Ctenophora as the sister lineage to other Metazoa. We find no support for the traditional view uniting the ctenophores and Cnidaria. Our findings are supported by Bayesian comparisons of topological hypotheses and we find no evidence that they are biased by long-branch attraction.

Conclusions: Our study further clarifies relationships among early branching metazoan lineages. Our phylogeny supports the still-controversial position of ctenophores as sister group to all other metazoans. This study also provides a workflow and computational tools for minimizing systematic bias in genome-based phylogenetic analyses. Future studies of metazoan phylogeny will benefit from ongoing efforts to sequence the genomes of additional invertebrate taxa that will continue to inform our view of the relationships among the major lineages of animals.


Fig. 1
Fig. 1
Visualization of gene ontology (GO) term enrichment across the total1080 dataset. In total, 142 GO terms were significantly enriched in the total1080 datasets compared to an outgroup reference annotation. Enrichments are depicted here for the Biological Process, Cellular Component and Molecular Function categories. In each, area subtended by a given GO term represents its frequency among significantly enriched GO terms
Fig. 2
Fig. 2
Distributions of long-branch scores and gene occupancy for Total1080 and Best108 matrices. In long-branch score heat maps, the scores were Z-scaled across columns to highlight among-taxon variability. Red indicates high long-branch scores relative to other taxa and blue denotes low scores. White in gene occupancy plots corresponds to missing data. The cladograms illustrate results of similarity by hierarchical clustering. Note that in the Total1080 dataset, Amphimedon, Mnemiopsis and Tetranychus cluster with other long-branched taxa that include the outgroups, the nematodes, and Oikopleura. However, in the Best108 matrix these taxa cluster with the main group, leaving only the outgroups, nematodes and Oikopleura in the long-branch cluster
Fig. 3
Fig. 3
Summary of phylogenetic results. Tree topology and branch lengths are derived from the Best108 matrix data set analysis under CAT-GTR. Support values represent: posterior probabilities from PhyloBayes [66] analysis of Best108 matrix under CAT-GTR/posterior probabilities from PhyloBayes analysis of MareMatrix matrix under CAT-GTR/bootstrap support in Total1080 matrix under a partitioned empirical model/108-locus jackknife from the 1080 locus set. Unannotated nodes have maximum support for all measures. Scale bar in substitutions per site. Silhouettes from For image attributions see Additional file 8
Fig. 4
Fig. 4
Summary of tests for Long Branch Attraction. Unrooted trees from analyses excluding putative long-branch taxa are shown. All analyses were conducted under maximum likelihood, partitioned empirical models. a Tree inferred without outgroups. b Tree inferred without outgroups and the sponge Amphimedon. c Tree inferred without the sponge Amphimedon. Black circles indicate bootstrap support of 100 % from 1000 replicates, red circles indicate support of 95 to 99 % and blue circles indicate support of 95 % or less. Non-bilaterian metazoan taxa are highlighted
Fig. 5
Fig. 5
Sensitivity analyses using progressive concatenation and rate binning. a Support for alternative phylogenetic hypotheses under progressive concatenation from the slowest evolving to the fastest evolving loci. The x-axis represents number of loci concatenated in order of rate of evolution, from 5 of the most slowly evolving at left to all 1080 loci at right. The y-axis indicates bootstrap support. Red circles in cladograms above corresponding plots denote the node for which bootstrap support was assessed. b Support for alternative phylogenetic hypotheses across the data. The x-axis represents bin number and the y-axis indicates bootstrap support. Bin number 1 contains 108 slowest evolving loci in the data set and bin number 10 contains 108 fastest evolving loci. Red circles in cladograms above corresponding plots denote the node for which bootstrap support was assessed

Similar articles

See all similar articles

Cited by 30 PubMed Central articles

See all "Cited by" articles


    1. Moroz LL, Kocot KM, Citarella MR, Dosung S, Norekian TP, Povolotskaya IS, et al. The ctenophore genome and the evolutionary origins of neural systems. Nature. 2014;510:109–114. doi: 10.1038/nature13400. - DOI - PMC - PubMed
    1. Bosch TC. Cnidarian-microbe interactions and the origin of innate immunity in metazoans. Annu Rev Microbiol. 2013;67:499–518. doi: 10.1146/annurev-micro-092412-155626. - DOI - PubMed
    1. Arendt D. The evolution of cell types in animals: emerging principles from molecular studies. Nat Rev Genet. 2008;9:868–882. doi: 10.1038/nrg2416. - DOI - PubMed
    1. Dunn CW, Hejnol A, Matus DQ, Pang K, Browne WE, Smith SA, et al. Broad phylogenomic sampling improves resolution of the animal tree of life. Nature. 2008;452:745–749. doi: 10.1038/nature06614. - DOI - PubMed
    1. Hejnol A, Obst M, Stamatakis A, Ott M, Rouse GW, Edgecombe GD, et al. Assessing the root of bilaterian animals with scalable phylogenomic methods. Proc Royal Soc B. 2015;276:4261–70. - PMC - PubMed

Publication types

LinkOut - more resources