2015-12-24

Ctenophora is not a sister group to other animals after all

Since 2008 there have been a debate whether the phylum Ctenophora is a phylogenetic sister group to other animals or not (Dunn et al. 2008; Philippe et al. 2009). There were some who claimed to have resolved the debate (that Ctenophora really is a sister group to other animals: Ryan et al. 2013), but the presented evidence was not at all convincing, as I discussed some time ago. Recently, another paper (Whelan et al. 2015) came out that favoured Ctenophora as the sister group to other animals (hereafter referred to as Ctenophora-basal tree), but this time the results were much more convincing and I was prepared to believe the Ctenophora-basal tree.

But not so fast. It turns out I was fooled. Only half a year after the Whelan et al. (2015) publication, Pisani et al. (2015) showed that the analyses by Whelan et al., as well as all previous analyses favouring Ctenophora-basal tree, were still plagued by systematic errors. Pisani et al. (2015) found that the sponges (Porifera) are probably the sister group to other animals (as is usually assumed) after all, although exact position of Ctenophora relative to the remaining animals still needs additional research. For me, one of the most striking results from Pisani et al. (2015) was the re-analysis of genome content data (presence/absence of protein coding genes) from Ryan et al. (2013). The genome content tree that Ryan et al. (2013) recovered, contained several phylogenetic relationships that are highly suspect (e.g. non-monophyly of Annelida and a group containing a mollusc, an annelid and a chordate Branchiostoma). Remarkably, when Pisani et al. (2015) used a less biased model of gene content evolution, which does not underestimate gene losses as much as did the model used by Ryan et al. (2013), they recovered an animal tree of life that fully agreed with independent phylogenomic analyses based gene sequences. There was not a single nonsense clade left. And in that tree, Porifera replaced Ctenophora as the sister group to other animals.

As for the Whelan et al. (2015) study that fooled me, the main issue comes down to the choice of an out-group. Whelan et al. (2015) analysed some of their datasets also with more realistic, but computationally more demanding CAT model, but recovered nevertheless Ctenophora-basal tree. It turns out, though, that when more distant out-groups to animals (Fungi or Fungi plus non-choanoflagllate members of Holozoa) are excluded from the analysis (retaining only the members of Holozoa or only the choanoflagllates), Whelan et al. (2015) datasets do not support Ctenophora-basal tree anymore or even favour Porifera-basal tree. Whelan et al. (2015) actually did analyse their datasets also with alternative out-group compositions, but only with simpler evolutionary models and therefore it was not possible to discover that different out-group compositions affect results when using more complex CAT model! Such a simple trick which the authors could have done to test their results...

Another recent study also found Ctenophora-basal tree (Chang et al. 2015) using the CAT model. Although Chang et al. (2015) analyses did not include Fungi, it was not tested if excluding more distant non-choanoflagllate Holozoa taxa could affect the results. But this Ctenophora question was not the topic of this paper (it was about Myxozoa, highly reduced parasitic Cnidaria), although one of the co-authors was (Hervé Philippe) also a co-author in Pisani et al. (2015) paper, which studied the effect of out-group composition.

Still, it is disconcerting that depending on the composition of the out-group, even if the closest relatives are included, the results can change that dramatically: using CAT model and including Fungi, the Ctenophora-basal tree can be recovered with maximal statistical support, but using CAT model and excluding all out-groups except choanoflagllates (the closest relatives of animals), Porifera-basal tree can be recovered with maximal support instead. I would have thought that as long as the closest out-groups are also included (choanoflagllates) in the analyses, the more distant out-groups (Fungi) should not influence the results that much. This tells me again (as I discussed before) that whatever phylogenetic signal there is for the relationships between Porifera, Ctenophora, Placozoa, Cnidaria, and Bilateria, it is quite tiny. These groups separated from each other in the Precambrian (>540 Ma) probably rather rapidly, perhaps within a few tens of millions of years or even within a shorter time period. Because it starts to look like the nervous systems and perhaps muscles might have evolved independently even three times in Ctenophora, Cnidaria, and Bilateria (Liebeskind et al. 2015; Moroz et al. 2014), it seems to be of less importance what are the exact phylogenetic relationships between Porifera, Ctenophora, Placozoa, Cnidaria, and Bilateria. Ctenophora, Cnidaria, and Bilateria might have evolved their morphological complexity independently from Porifera or Placozoa-like ancestors.

References

Chang ES, Neuhof M, Rubinstein ND, Diamant A, Philippe H, Huchon D, Cartwright P (2015) Genomic insights into the evolutionary origin of Myxozoa within Cnidaria. Proceedings of the National Academy of Sciences 112: 14912–14917. doi: 10.1073/pnas.1511468112
Dunn CW, Hejnol A, Matus DQ, Pang K, Browne WE, Smith S a, Seaver E, Rouse GW, Obst M, Edgecombe GD, Sørensen M V, Haddock SHD, Schmidt-Rhaesa A, Okusu A, Kristensen RM, Wheeler WC, Martindale MQ, Giribet G (2008) Broad phylogenomic sampling improves resolution of the animal tree of life. Nature 452: 745–749. doi: 10.1038/nature06614
Liebeskind BJ, Hillis DM, Zakon HH (2015) Convergence of ion channel genome content in early animal evolution. Proceedings of the National Academy of Sciences 112: E846–E851. doi: 10.1073/pnas.1501195112
Moroz LL, Kocot KM, Citarella MR, Dosung S, Norekian TP, Povolotskaya IS, Grigorenko AP, Dailey C, Berezikov E, Buckley KM, Ptitsyn A, Reshetov D, Mukherjee K, Moroz TP, Bobkova Y, Yu F, Kapitonov V V, Jurka J, Bobkov Y V, Swore JJ, Girardo DO, Fodor A, Gusev F, Sanford R, Bruders R, Kittler E, Mills CE, Rast JP, Derelle R, Solovyev V V, Kondrashov F a, Swalla BJ, Sweedler J V, Rogaev EI, Halanych KM, Kohn AB (2014) The ctenophore genome and the evolutionary origins of neural systems. Nature 510: 109–114. doi: 10.1038/nature13400
Philippe H, Derelle R, Lopez P, Pick K, Borchiellini C, Boury-Esnault N, Vacelet J, Renard E, Houliston E, Quéinnec E, Da Silva C, Wincker P, Le Guyader H, Leys S, Jackson DJ, Schreiber F, Erpenbeck D, Morgenstern B, Wörheide G, Manuel M (2009) Phylogenomics revives traditional views on deep animal relationships. Current Biology 19: 706–712. doi: 10.1016/j.cub.2009.02.052
Pisani D, Pett W, Dohrmann M, Feuda R, Rota-Stabelli O, Philippe H, Lartillot N, Wörheide G (2015) Genomic data do not support comb jellies as the sister group to all other animals. Proceedings of the National Academy of Sciences 112: 15402–15407. doi: 10.1073/pnas.1518127112
Ryan JF, Pang K, Schnitzler CE, Nguyen A-D, Moreland RT, Simmons DK, Koch BJ, Francis WR, Havlak P, Smith S a, Putnam NH, Haddock SHD, Dunn CW, Wolfsberg TG, Mullikin JC, Martindale MQ, Baxevanis AD (2013) The Genome of the Ctenophore Mnemiopsis leidyi and Its Implications for Cell Type Evolution. Science 342: 1242592–1242592. doi: 10.1126/science.1242592
Whelan N V., Kocot KM, Moroz LL, Halanych KM (2015) Error, signal, and the placement of Ctenophora sister to all other animals. Proceedings of the National Academy of Sciences 112: 5773–5778. doi: 10.1073/pnas.1503453112

2 comments:

  1. Hey Marko. I thought this was great summary.
    Thought you might be interested in this paper I found which directly refutes one of Ryan's (2013) claims - that major evidence supporting the Ctenophore-first hypothesis is Ctenophora's lack of HOX genes. Apparently Cnidaria doesn't have true HOX genes either:

    Kamm, K., Schierwater, B., Jakob, W., Dellaporta, S. L., & Miller, D. J. (2006). Axial patterning and diversification in the cnidaria predate the Hox system. Current Biology, 16(9), 920-926.

    Cheers.
    Melanie

    ReplyDelete
  2. Hey Melanie, thank you for pointing to this paper. I had a pdf of that and maybe had even read it, but completely forgotten. I found few more papers with similar conclusion: http://dx.doi.org/10.1371/journal.pone.0004231
    http://dx.doi.org/10.1016/j.mod.2015.08.005

    I guess it's partly matter of definition which are "real" HOX genes and which not, but it looks like that there are functionally significant differences between Cnidaria and Bilateria and that there have been independent expansions of these genes within Cnidaria. Seems to be a small additional support for the idea that morphological complexity evolved independently in Cnidaria, Bilateria and Ctenophora.

    ReplyDelete