Four myriapod relatives – but who are sisters? No end to debates on relationships among the four major myriapod subgroups

Abstract

Background Phylogenetic relationships among the myriapod subgroups Chilopoda, Diplopoda, Symphyla and Pauropoda are still not robustly resolved. The first phylogenomic study covering all subgroups resolved phylogenetic relationships congruently to morphological evidence but is in conflict with most previously published phylogenetic trees based on diverse molecular data. Outgroup choice and long-branch attraction effects were stated as possible explanations for these incongruencies. In this study, we addressed these issues by extending the myriapod and outgroup taxon sampling using transcriptome data. Results We generated new transcriptome data of 42 panarthropod species, including all four myriapod subgroups and additional outgroup taxa. Our taxon sampling was complemented by published transcriptome and genome data resulting in a supermatrix covering 59 species. We compiled two data sets, the first with a full coverage of genes per species (292 single-copy protein-coding genes), the second with a less stringent coverage (988 genes). We inferred phylogenetic relationships among myriapods using different data types, tree inference, and quartet computation approaches. Our results unambiguously support monophyletic Mandibulata and Myriapoda. Our analyses clearly showed that there is strong signal for a single unrooted topology, but a sensitivity of the position of the internal root on the choice of outgroups. However, we observe strong evidence for a clade Pauropoda+Symphyla, as well as for a clade Chilopoda+Diplopoda. Conclusions Our best quartet topology is incongruent with current morphological phylogenies which were supported in another phylogenomic study. AU tests and quartet mapping reject the quartet topology congruent to trees inferred with morphological characters. Moreover, quartet mapping shows that confounding signal present in the data set is sufficient to explain the weak signal for the quartet topology derived from morphological characters. Although outgroup choice affects results, our study could narrow possible trees to derivatives of a single quartet topology. For highly disputed relationships, we propose to apply a series of tests (AU and quartet mapping), since results of such tests allow to narrow down possible relationships and to rule out confounding signal.

What Is an “Arachnid”? Consensus, Consilience, and Confirmation Bias in the Phylogenetics of Chelicerata

“…Most of the activities of this school are no different from what molecular phylogeneticists simply call molecular dating, which is a widely implemented technique. Some of the recurring controversies engendered by "molecular paleobiologists" have stemmed from overconfidence in the completeness of the fossil record, overconfidence in morphological phylogenies (or specific, arbitrarily selected morphological phylogenies, when competing hypotheses are available), overvaluing morphological and paleontological evidence in the assessment of competing molecular hypotheses, overinterpreting paleontological and morphological data in establishing calibrations in molecular dating; outright misconstruing or misrepresenting the morphological literature to accord with a preferred molecular topology; and consistently ignoring the repeated observation by other research groups that site heterogeneous models (as well as other proposed silver bullets to achieving morphological trees with molecular data) do not actually achieve the desired result across datasets [10,44,[54][55][56][57][58][59]. However, given that morphological hypotheses of relationships predate molecular counterparts by decades or sometimes centuries, "molecular paleobiological" results are often unguardedly accepted by the broader community for their palatability, specifically by those who do not examine the underlying phylogenomic data and analyses or lack the expertise to do so.…”

A Rearrangement of the Mitochondrial Genes of Centipedes (Arthropoda, Myriapoda) with a Phylogenetic Analysis

“…However, recent phylogenomic analyses based on transcriptomic data showed slightly different results. The results obtained by Szucsich et al, Benavides et al, and Wang et al are basically the same, demonstrating strong evidence for the clade Pauropoda + Symphyla (=Edafopoda) as well as for Chilopoda + Diplopoda (=Pectinopoda) [ 19 , 21 , 22 ]. The difference is that Fernández et al identified two alternative phylogenetic relationships for Symphyla: one that classifies it as a sister group to the Diplopoda + Chilopoda, and one that places Symphyla closer to Dignatha [ 27 ].…”

“…In some studies, Strigamia maritima is treated as an ideal model species for ecological and developmental research [ 4 , 5 ]. Recent studies using either comparative morphological or molecular evidence have found that myriapods and all extant myriapod classes are monophyletic [ 6 , 7 , 8 , 9 , 10 , 11 , 12 , 13 , 14 , 15 , 16 , 17 , 18 , 19 , 20 ]. Most of these analyses support the contention that the Chilopoda represent the basal lineage of the Myriapoda, with the remaining three classes united as the Progoneata.…”

Are fleas highly modified Mecoptera? Phylogenomic resolution of Antliophora (Insecta: Holometabola)

“…We defined four lineages: Nannochoristidae, Boreidae, Pistillifera, and Siphonaptera. This arrangement did not allow us to test for a monophyly of Mecoptera due to the unrooted nature of the resulting quartets (see also Szucsich et al, 2020 for a similar rooting issue in myriapod relationships). However, it enabled us to re-examine varying signal for the phylogenetic relationship between Nannochoristidae and Siphonaptera on one hand and that between Boreidae and Siphonaptera on the other (as hypothesized by Whiting 2002).…”