CoChat
Sign Up
Cited by 154Protist diversity and phylogenyGenomics and Phylogenetic StudiesRNA and protein synthesis mechanisms

Phylogeny of eukaryotes based on ribosomal RNA: Long-branch attraction and models of sequence evolution

Philippe* and, Hervé, Germot, Agnès·2000·Molecular Biology and Evolution
DOI 10.1093/oxfordjournals.molbev.a026362
Open this paper in CoChat — free account

Abstract

The molecular phylogeny of eukaryotes was first based on the comparison of the small-subunit ribosomal RNA (SSU rRNA) and quickly became a paradigm (Sogin 1991 ). However, serious criticisms have recently been raised against it. In particular, it has been suggested that the various protist lineages that emerged first (e.g., microsporidia, trichomonads, mycetozoans) were misplaced because of the long-branch attraction (LBA) artifact (Philippe and Adoutte 1998 ; Philippe and Laurent 1998 ; Germot and Philippe 1999 ; Moreira, Le Guyader, and Philippe 1999 ; Roger et al. 1999 ; Stiller and Hall 1999 ). Indeed, simulation studies (unpublished data) showed that when a distant outgroup was used, all the standard methods of phylogenetic reconstruction artifactually inferred a phylogeny with an asymmetrical base containing the fast-evolving lineages, leading to a tree shape very similar to the rRNA one. When the substitution rates are considerably variable for different lineages, the use of a model that reflects the actual substitution probability is of utmost importance for obtaining a correct tree topology (Lockhart et al. 1994 ; Swofford et al. 1996 ; Yang 1996 ; Takezaki and Gojobori 1999 ). In the case of rRNA-based eukaryotic phylogeny, the first model violation noticed was the variation of G + C content among species (Hasegawa and Hashimoto 1993 ). Yet, the overall picture of the rRNA tree remained unchanged when only species with similar G + C contents were included (Leipe et al. 1993 ) or when methods less sensitive to this bias, such as transversion parsimony (Woese et al. 1991 ) or log-Det distance (Lockhart et al. 1994 ), were applied. Variation in evolutionary rates across sites (RAS) is a further complexity of sequence evolution which also has a significant effect on evolutionary tree building (Olsen 1987 ; Yang 1996 ). For instance, the late emergence of fast-evolving lineages such as the nucleomorph of chloroarachniophytes (Van de Peer et al. 1996 ) or microsporidia (Peyretaillade et al. 1998 ) can be retrieved only when RAS models are assumed. Finally, the existence of covarion structure constitutes a further model violation that has just begun to be considered (Lockhart et al. 1998 ; Lopez, Forterre, and Philippe 1999 ). In this study, we sequenced the complete large-subunit (LSU) rRNA of Trichomonas vaginalis. We concatenated this sequence with that of the SSU rRNA for this taxon and compared these data with similar data from other eukaryotes. Using these data, we investigated the robustness of eukaryotic phylogeny with consideration for the model of sequence evolution used. The protocol for obtaining the LSU rRNA using several universal and specific primers was the same as that previously described (Germot, Philippe, and Le Guyader 1996 ), except that we sequenced both strands here. The size of the LSU rRNA was approximately 2,760 nt (the beginning and end were identified only by homology with other rRNAs), and its G + C content was 46%. The sequence of T. vaginalis was aligned manually with the available complete LSU rRNA sequences of 120 eukaryotes and of 15 archaebacteria, using the ED program of the MUST package (Philippe 1993 ). To increase the sequence lengths, we aligned SSU rRNA sequences for the same set of organisms whenever possible, and only the fusion of SSU and LSU rRNA was used for further analysis. A subset of only 32 species, encompassing as much biodiversity as possible (i.e., by eliminating numerous fungal, plant, and metazoan species), was used in order to reduce the computing time of our maximum-likelihood (ML) analysis. Only positions that could be unambiguously aligned were selected, leaving 875 and 1,604 positions for SSU and LSU rRNA, respectively. The alignment is available on request or at http://bufo.bc4.u-psud.fr. The G + C content of rRNA appeared to be significantly heterogeneous between species at the 5% level, as measured by the χ2 statistics implemented in PUZZLE (Strimmer and von Haeseler 1996 ). Rather than coping with this model violation using a nonhomogeneous model (e.g., Galtier and Gouy 1998 ), only transversions were considered, since the ratio of purines to pyrimidines appeared to be homogeneous for all of the considered organisms, as previously observed for prokaryotic rRNA (Woese et al. 1991 ). This allowed us to use complex RAS models, such as those using a Γ law distribution. Moreover, transitions were more saturated than transversions in a dot plot comparison (data not shown) and therefore were less likely to contain a reliable ancient phylogenetic signal. The ML tree that was inferred with a model assuming equal rates across sites was very similar to the ones inferred from the SSU rRNA only (fig. 1A ). The three “archezoan” lineages (trichomonads, microsporidia, and diplomonads) emerged first, followed by euglenozoans and mycetozoans, and then by the crown. Within the tree, the monophyly of the various phyla was recovered with good statistical supp

Integrity signals

Registry

Crossref · 10.1093/oxfordjournals.molbev.a026362

Cites

34 of 34 references verified0 citation errors · 0 not checkable

Statistics

Recomputing p-values needs the paper’s full text.

Claims the paper makes

This paper’s full text isn’t openly available, so its claims and self-check weren’t run. The registry and citation checks still apply.

Claims other papers make about this one

Each is checked against this paper’s own text: does it say what the citing paper says it does?

  • Inclusion of outgroups very distant from the ingroup can cause reconstruction artifacts by attracting fast-evolving (long-branched) ingroup species toward the root.

    “the inclusion of outgroups very distant from the ingroup can cause reconstruction artifacts by attracting fast-evolving (long-branched) ingroup species toward the root (25, 27 -31)”

    From Genomic data do not support comb jellies as the sister group to all other animals (2015) · cites this paper as [28]

Cites

Flagged first, then by citations

1 of the 35 references couldn’t be matched to a known paper.

Cited by

Sorted by citations

Related

Authors
Philippe* and, H., & Germot, A.
ORCID
Germot, Agnès
Venue
Molecular Biology and Evolution
Publisher
Oxford University Press (OUP)
Published
2000-05-01
Volume
17
Issue
5
Pages
830-834
ISSN
  • 0737-4038 (ISSN-L)
  • 1537-1719
DOI
10.1093/oxfordjournals.molbev.a026362
OpenAlex
W2117688872
Cited by
154
References
35
Citations by year
2025: 1 · 2022: 2 · 2019: 2 · 2018: 3 · 2017: 6 · 2016: 2 · 2015: 4 · 2014: 4 · 2013: 6 · 2012: 8
FWCI
5.17
Type
journal article
Language
en
Open access
closed
Locations
  • Molecular Biology and Evolution (publishedVersion)
  • PubMed (publishedVersion)
  • HAL (submittedVersion)
Topics
  • Life Sciences › Biochemistry, Genetics and Molecular Biology › Molecular Biology › Protist diversity and phylogeny
  • Life Sciences › Biochemistry, Genetics and Molecular Biology › Molecular Biology › Genomics and Phylogenetic Studies
  • Life Sciences › Biochemistry, Genetics and Molecular Biology › Molecular Biology › RNA and protein synthesis mechanisms
Keywords
eukaryotic phylogeny, small subunit rRNA, long-branch attraction, protist phylogeny, microsporidia, trichomonads, nucleomorph, chlorarachniophytes, substitution rate heterogeneity, GC content, outgroup selection, sequence evolution models
MeSH terms
Animals, Base Sequence, Humans, Models, Genetic, Molecular Sequence Data, Phylogeny, RNA, Ribosomal, Likelihood Functions, Evolution, Molecular

About these checks

Registry, retraction, Cites and statistics results are computed, not judged, with one exception: a printed reference that no index lists is read once by a model before it counts as a citation error. Claims are extracted by a model and kept only when their quote appears verbatim in the text; each cited source is then checked, with a model’s help, for whether it says what the claim relies on. Those verdicts can be wrong — read the claim’s quote and the cited source before relying on one.

Statistics and arithmetic are recomputed, not judged.

Confidence bands are shown only for fields calibrated on our annotated test set. None is calibrated yet, so no band is shown.

What counts as a claim ›How we measure confidence ›How we check this ›

Are you an author, or do you see a problem with this page? Request a review