Search

Phylogenetic analyses may suffer from multiple sources of error leading to conflict between genes and methods of inference. The evolutionary history of the mollusc clade Vetigastropoda makes them susceptible to these conflicts, their higher level phylogeny remaining largely unresolved. Originating over 350 Ma, vetigastropods were the dominant marine snails in the Paleozoic. Multiple extinction events and new radiations have resulted in both very long and very short branches and a large extant diversity of over 4000 species. This is the perfect setting of a hard phylogenetic question in which sources of conflict can be explored. We present 41 new transcriptomes across the diversity of vetigastropods (62 terminals total), and provide the first genomic-scale phylogeny for the group. We find that deep divergences differ from previous studies in which long branch attraction was likely pervasive. Robust results leading to changes in taxonomy include the paraphyly of the order Lepetellida and the family Tegulidae. Tectinae subfam. nov. is designated for the clade comprising Tectus, Cittarium, and Rochia. For two early divergences, topologies disagreed between concatenated analyses using site heterogeneous models versus concatenated partitioned analyses and summary coalescent methods. We investigated rate and composition heterogeneity among genes, as well as missing data by locus and by taxon, none of which had an impact on the inferred topologies. We also found no evidence for ancient introgression throughout the phylogeny. We further tested whether uninformative genes and over-partitioning were responsible for this discordance by evaluating the phylogenetic signal of individual genes using likelihood mapping, and by analyzing the most informative genes with a full multispecies coalescent (MSC) model. We find that most genes are not informative at the two conflicting nodes, but neither this nor gene-wise partitioning are the cause of discordant results. New method implementations that simultaneously integrate amino acid profile mixture models and the MSC might be necessary to resolve these and other recalcitrant nodes in the Tree of Life. [Fissurellidae; Haliotidae; likelihood mapping; multispecies coalescent; phylogenetic signal; phylogenomic conflict; site heterogeneity; Trochoidea.]

Long branch attraction (LBA) is a prevalent form of bias in phylogenetic estimation but the reasons for it are only partially understood. We argue here that it is largely due to differences in the sizes of the model spaces corresponding to different trees. Trees with long branches together allow much more flexible internal branch length parameter estimation. Consequently, although each tree has the same number of parameters, trees with long branches together have larger effective model spaces. The problem of LBA becomes particularly pronounced with partitioned data. Formulation of tree estimation as model selection leads us to propose bootstrap bias corrections as cross-checks on estimation when long branches end up being estimated together. [Bootstrap; long branch attraction; maximum likelihood; model selection; partitioned model; phylogenetics.]

Long-branch attraction is a systematic artifact that results in erroneous groupings of fast-evolving taxa. The combination of short, deep internodes in tandem with long-branch attraction artifacts has produced empirically intractable parts of the Tree of Life. One such group is the arthropod subphylum Chelicerata, whose backbone phylogeny has remained unstable despite improvements in phylogenetic methods and genome-scale data sets. Pseudoscorpion placement is particularly variable across data sets and analytical frameworks, with this group either clustering with other long-branch orders or with Arachnopulmonata (scorpions and tetrapulmonates). To surmount long-branch attraction, we investigated the effect of taxonomic sampling via sequential deletion of basally branching pseudoscorpion superfamilies, as well as varying gene occupancy thresholds in supermatrices. We show that concatenated supermatrices and coalescent-based summary species tree approaches support a sister group relationship of pseudoscorpions and scorpions, when more of the basally branching taxa are sampled. Matrix completeness had demonstrably less influence on tree topology. As an external arbiter of phylogenetic placement, we leveraged the recent discovery of an ancient genome duplication in the common ancestor of Arachnopulmonata as a litmus test for competing hypotheses of pseudoscorpion relationships. We generated a high-quality developmental transcriptome and the first genome for pseudoscorpions to assess the incidence of arachnopulmonate-specific duplications (e.g., homeobox genes and miRNAs). Our results support the inclusion of pseudoscorpions in Arachnopulmonata (new definition), as the sister group of scorpions. Panscorpiones (new name) is proposed for the clade uniting Scorpiones and Pseudoscorpiones.

Biochemical demands constrain the range of amino acids acceptable at specific sites resulting in across-site compositional heterogeneity of the amino acid replacement process. Phylogenetic models that disregard this heterogeneity are prone to systematic errors, which can lead to severe long branch attraction artifacts. State-of-the-art models accounting for across-site compositional heterogeneity include the CAT model, which is computationally expensive, and empirical distribution mixture models estimated via maximum likelihood (C10 to C60 models). Here, we present a new, scalable method EDCluster for finding empirical distribution mixture models involving a simple cluster analysis. The cluster analysis utilizes specific coordinate transformations which allow the detection of specialized amino acid distributions either from curated databases, or from the alignment at hand. We apply EDCluster to the HOGENOM and HSSP databases in order to provide universal distribution mixture (UDM) models comprising up to 4096 components. Detailed analyses of the UDM models demonstrate the removal of various long branch attraction artifacts and improved performance compared to the C10 to C60 models. Ready-to-use implementations of the UDM models are provided for three established software packages (IQ-TREE, Phylobayes, and RevBayes).

Fabales is a cosmopolitan angiosperm order that consists of four families: Leguminosae (Fabaceae), Polygalaceae, Surianaceae, and Quillajaceae. Despite the great interest in this group, a convincing phylogeny of the order is still not available. Therefore, the aim of this study was to explicitly test for possible long branch attraction (LBA) problems within Fabales for the first time, and determine whether low tree stemminess and unequal branch lengths could worsen this problem. Supermatrix analysis of Fabales was carried out using previously published plastid matK, trnL, rbcL, and newly sequenced nuclear sqd1 regions for 678 taxa in total, including 43 outgroup taxa from families of Fabidae. We employed additional analyses, such as simulations, network analyses, sampling different outgroup taxa (random or real), removing fast evolving sites and fast evolving taxa, and molecular clock rooting, to identify both LBA and (or) rooting problems. These analyses clearly show that the Fabales phylogeny has been influenced by the sampling of outgroup taxa, but not LBA. However, network analyses show that even though it is weak, there is a consistent phylogenetic signal among the rapidly radiated Fabales families, which can be traced by further analyses. While, molecular clock rooting analysis yielded a (Leguminosae(Polygalaceae(Surianaceae+Quillajaceae))) topology with strong support for the first time here, supermatrix analyses yielded a ((Leguminosae+Polygalaceae)(Surianaceae+Quillajaceae)) with low-moderate support. © 2020, Canadian Science Publishing. All rights reserved.

Due to morphological reduction and absence of amplifiable plastid genes, the identification of photosynthetic relatives of heterotrophic plants is problematic. Although nuclear and mitochondrial gene sequences may offer a welcome alternative source of phylogenetic markers, the presence of rate heterogeneity in these genes may introduce bias/systematic error in phylogenetic analyses. We examine the phylogenetic position of Thismiaceae based on nuclear 18S rDNA and mitochondrial atpA DNA sequence data, as well as using parsimony, likelihood and Bayesian inference methods. Significant differences in evolutionary rates of these genes between closely related taxa lead to conflicting results: while parsimony analyses of 18S rDNA and combined data strongly support the monophyly of Thismiaceae, Bayesian inference, with and without a relaxed molecular clock, as well as the Swofford-Olsen-Waddell-Hillis (SOWH) test confidently reject this hypothesis. We show that rate heterogeneity in our data leads to long-branch attraction artifacts in parsimony analysis. However, using model-based inference methods the question of whether Thismiaceae are monophyletic remains elusive. On the one hand maximum likelihood nonparametric bootstrapping and parametric hypothesis tests fail to support a paraphyletic Thismiaceae, on the other hand Bayesian inference methods (both without and with a relaxed clock) significantly reject a monophyletic Thismiaceae. These results show that an adequate sampling, the use of rate homogeneous data, and the application of different inference methods are important factors for developing phylogenetic hypotheses of myco-heterotrophic plants. (C) The Willi Hennig Society 2009. ispartof: Cladistics vol:25 issue:1 pages:64-77 ispartof: location:United States status: published

It is commonly assumed that increasing the number of characters has the potential to resolve evolutionary radiations. Here, we studied photosynthetic stramenopiles (Ochrophyta) using alignments of heterogeneous origin mitochondrion, plastid, and nucleus. Surprisingly while statistical support for the relationships between the six major Ochrophyta lineages increases when comparing the mitochondrion (6,762 sites) and plastid (21,692 sites) trees, it decreases in the nuclear (209,105 sites) tree. Statistical support is not simply related to the data set size but also to the quantity of phylogenetic signal available at each position and our ability to extract it. Here, we show that this ability for current phylogenetic methods is limited, because conflicting results were obtained when varying taxon sampling. Even though the use of a better fitting model improved signal extraction and reduced the observed conflicts, the plastid data set provided higher statistical support for the ochrophyte radiation than the larger nucleus data set. We propose that the higher support observed in the plastid tree is due to an acceleration of the evolutionary rate in one short deep internal branch, implying that more phylogenetic signal per position is available to resolve the Ochrophyta radiation in the plastid than in the nuclear data set. Our work therefore suggests that, in order to resolve radiations, beyond the obvious use of data sets with more positions, we need to continue developing models of sequence evolution that better extract the phylogenetic signal and design methods to search for genes/characters that contain more signal specifically for short internal branches.

International audience; Phylogenomics seeks to use next‐generation data to robustly infer an organism's evolutionary history. Yet, the practical caveats of phylogenomics motivate investigation of improved efficiency, particularly when quality of phylogenies are questionable. To achieve improvements, one goal is to maintain or enhance the quality of phylogenetic inference while severely reducing dataset size. We approach this by assessing which kinds of loci in phylogenomic alignments provide the majority of support for a phylogenetic inference of cockroaches in Blaberoidea. We examine locus substitution rate, saturation, evolutionary divergence, rate heterogeneity, stabilizing selection, and a priori information content as traits that may determine optimality. Our controlled experimental design is based on 265 loci for 102 blaberoidean taxa and 22 outgroup species. Loci with high substitution rate, low saturation, low sequence distance, low rate heterogeneity, and strong stabilizing selection derive more support for phylogenetic relationships. We found that some phylogenetic information content estimators may not be meaningful for assessing information content a priori. We use these findings to design concatenated datasets with an optimized subsample of 100 loci. The tree inferred from the optimized subsample alignment was largely identical to that inferred from all 265 loci but with less evidence of long branch attraction, improved statistical support, and potential 4‐6x improvements to computation time. Supported by phylogenetic and morphological evidence, we erect three newly named clades (Anallactinae Evangelista & Wipfler subfam. nov., Orkrasomeria tax. nov. Evangelista, Wipfler, & Béthoux and Hemithyrsocerini Evangelista tribe nov.) and propose other taxonomic modifications. The diagnosis of Pseudophyllodromiidae Grandcolas, 1996 is modified to accommodate Anallactinae and Pseudophyllodromiinae Vickery & Kevan, 1983. The diagnosis of Ectobiidae Brunner von Wattenwyl, 1865 is modified to add novel morphological characters.

Chaetonotidae is the most diverse and widely distributed family of the order Chaetonotida (Gastrotricha) and includes both marine and freshwater species. Although the family is regarded as a sister taxon to the exclusively marine Xenotrichulidae, the type of environment, marine or freshwater, where Chaetonotidae originated is still not known. Here, we reconstructed the phylogeny of the family based on molecular sequence data and mapped both morphological and ecological characters to determine the ancestral environment of the first members of the family. Our results revealed that the freshwater genus Bifidochaetus is the earliest branching lineage in the paraphyletic Chaetonotidae (encompassing Dasydytidae and Neogosseidae). Moreover, we reconstructed Lepidochaetus-Cephalionotus clade as a monophyletic sister group to the remaining chaetonotids, which supports Kisielewski's morphological based hypothesis concerning undifferentiated type of body scales as a most primary character in Chaetonotidae. We also found that reversals to marine habitats occurred independently in different Chaetonotidae lineages, thus marine species in the genera Heterolepidoderma, Halichaetonotus, Aspidiophorus and subgenera Chaetonotus (Schizochaetonotus) or Chaetonotus (Marinochaetus) should be assumed as having secondarily invaded the marine environment. Character mapping revealed a series of synapomorphies that define the clade that includes Chaetonotidae (with Dasydytidae and Neogosseidae), the most important of which may be those linked to reproduction.

Summary Species of the family Tylenchidae are encountered in large numbers in soils. The genus Labrys was recently described characterised by a remarkable lip pattern that differs from all other known Tylenchidae genera. Here we describe a curious new species, Labrys fujianensis sp. n., that morphologically fits the genus Labrys but which is genetically divergent. The phylogeny was inferred based on 18S and 28S rDNA and light and scanning electron microscopy were used to extract detailed morphologies. The phylogenetic position of this species and its phenotypic convergence are discussed. The possibility of a long-branch attraction artefact was inspected both by removal of variable nucleotide sites and monophyletic testing of topologies. The results confirmed the divergent positioning of the presented species and it is demonstrated that the genetic diversity in Tylenchidae may be much higher than expected due to morphological homoplasy.

Despite their ecological and evolutionary importance as key components of tropical ecosystems, the phylogeny of fig trees is still unresolved. We use restriction-site-associated DNA (RAD) sequencing (ca420kb) and 102 morphological characters to elucidate the relationships between 70 species ofFicusrepresenting all known subgenera and sections and five outgroups. We compare morphological and molecular results to highlight discrepancies and reveal possible inference bias. We analyse marker and taxon properties that may bias molecular inferences, with existing softwares and a new approach based on iterative principal component analysis to reduce variance between clusters of samples. For the first time, with both molecular and morphological data, we recover a monophyletic subgenusUrostigmaand a clade with all gynodioecious fig trees. However, our analyses show that it is not possible to homogenize evolutionary rates and GC content for all taxa prior to phylogenetic inference and that four competing positions for the root of the molecular tree are possible. The placement of the long-branched sectionPharmacosyceaas sister to all other fig trees is not supported by morphological data and considered as a result of a long branch attraction artefact to the outgroups. Regarding morphological features and indirect evidence from the pollinator tree of life, the topology that divides the genusFicusinto monoeciousversusgynodioecious species appears most likely. Active pollination is inferred as the ancestral state for all topologies, ambiguity remains for ancestral breeding system including for the favored topology, and it appears most likely that the ancestor of fig trees was a freestanding tree. Increasing sampling may improve results and would be at least as relevant as maximizing the number of sequenced regions given the strong heterogeneity in evolutionary rates, and to a lesser extent, base composition among species. Despite morphological plasticity and frequent homoplasy of multiple characters, we advocate giving a central role to morphology in our understanding of the evolution ofFicus, especially as it can help detect insidious systematic errors that tend to become more pronounced with larger molecular data sets.

Assessing support for molecular phylogenies is difficult because the data is heterogeneous in quality and overwhelming in quantity. Traditionally, node support values (bootstrap frequency, Bayesian posterior probability) are used to assess confidence in tree topologies. Other analyses to assess the quality of phylogenetic data (e.g. Lento plots, saturation plots, trait consistency) and the resulting phylogenetic trees (e.g. internode certainty, parameter permutation tests, topological tests) exist but are rarely applied. Here we argue that a single qualitative analysis is insufficient to assess support of a phylogenetic hypothesis and relate data quality to tree quality. We use six molecular markers to infer the phylogeny of Blattodea and apply various tests to assess relationship support, locus quality, and the relationship between the two. We use internode-certainty calculations in conjunction with bootstrap scores, alignment permutations, and an approximately unbiased (AU) test to assess if the molecular data unambiguously support the phylogenetic relationships found. Our results show higher support for the position of Lamproblattidae, high support for the termite phylogeny, and low support for the position of Anaplectidae, Corydioidea and phylogeny of Blaberoidea. We use Lento plots in conjunction with mutation-saturation plots, calculations of locus homoplasy to assess locus quality, identify long branch attraction, and decide if the tree's relationships are the result of data biases. We conclude that multiple tests and metrics need to be taken into account to assess tree support and data robustness.

Catfishes (Teleostei: Siluriformes) are a highly diverse order within Ostariophysi that is distributed worldwide. At the base of this clade emerge three lineages with well-defined monophylies: Diplomystidae, Loricarioidei, and Siluroidei. Morphological phylogeny studies place the Diplomystidae as the earliest branching of these three lineages, but studies based on molecular phylogenetics consistently find the fast-evolving Loricarioidei instead. The high lineage evolutionary rate heterogeneity in this order and the fact that the lineage placed closest to the root in the molecular phylogenies is fast evolving, including many long branches, raises the possibility that the discrepancy between morphological and molecular phylogenies may be the result of a long branch attraction inference artifact. We test this hypothesis by using a 10-gene dataset to evaluate the arrangement of the three main siluriform lineages, and apply the LS3 and LS4 taxon sequence subsampling methods to reduce evolutionary rate heterogeneity among lineages. The initial and complete dataset supports the basal branching of Loricarioidei as in all previous molecular phylogenies, but once lineage rate heterogeneity is reduced with LS3 or LS4 through the removal of sequences disrupting homogeneity, the phylogeny shows Diplomystidae as the earliest branching group, with high supports, as proposed by morphological phylogeny. The result obtained with LS3, however, introduces the misplacement of one of the species with the highest amount of missing data, Scoloplax sp. Because the sequence sub-selection criterion of LS4 has been optimized to reduce data removal, the phylogeny resulting from the LS4-processed data is in agreement with the known intra-lineage relationships in addition to supporting the morphologically-based rooting hypothesis. Our results are the first instance in which a consensus between molecular and morphological phylogeny is reached concerning the root of this order.

The Guianas are one of the most diverse regions of the Neotropics, hosting a particularly high rate of freshwater fish endemism. The present distributional patterns of freshwater fish species in the major catchments of the Guianas (comprising Guyana, Suriname and French Guiana) were analysed to reveal the faunal relationships between rivers, evaluate different hypotheses concerning biogeographical units, and redefine the boundaries of the Guianese freshwater ecoregions. A parsimony analysis of endemicity was performed using a data partitioning strategy to alleviate some drawbacks inherent to the method (e.g. long branch attraction artefact, heterotachy), and take into account alternative parsimony models assigning different constraints on state changes for the different species. A strong spatial element was present in the data with a structuring of species along a west-east gradient. Two main biogeographical units were highlighted: one to the west, ranging from the Essequibo to the Commewijne rivers and including the Proto Berbice and Surinamese regions, and one to the east ranging from the Maroni to the Oyapock rivers and including the Western, Central and Eastern French Guiana regions. Each ecoregion possessed distinctive fish assemblages, and three to four potential zones of faunal exchanges between Amazonian and Guianese rivers have been confirmed.

Premise of the study Phylogenomic studies employing large numbers of genes, including those based on plastid genomes (plastomes), are becoming common. Nonphotosynthetic plants such as mycoheterotrophs (which rely on root-associated fungi for essential nutrients, including carbon) tend to have highly elevated rates of plastome evolution, substantial genome reduction, or both. Mycoheterotroph plastomes therefore provide excellent test cases for investigating how extreme conditions impact phylogenomic inference. Methods We used parsimony and likelihood analysis of protein-coding gene sets from published and newly completed plastomes to infer the phylogenetic placement of taxa from the 10 angiosperm families in which mycoheterotrophy evolved. Key results Despite multiple very long branches that reflect elevated substitution rates, and frequently patchy gene recovery due to genome reduction, inferred phylogenetic placements of most mycoheterotrophic lineages in DNA-based likelihood analyses are both well supported and congruent with other studies. Amino-acid-based likelihood placements are broadly consistent with DNA-based inferences, but extremely rate-elevated taxa can have unexpected placements-albeit with weak support. In contrast, parsimony analysis is strongly misled by long-branch attraction among many distantly related mycoheterotrophic monocots. Conclusions Mycoheterotrophic plastomes provide challenging cases for phylogenomic inference, as substitutional rates can be elevated and genome reduction can lead to sparse gene recovery. Nonetheless, diverse likelihood frameworks provide generally well-supported and mutually concordant phylogenetic placements of mycoheterotrophs, consistent with recent phylogenetic studies and angiosperm-wide classifications. Previous predictions of parallel photosynthesis loss within families are supported for Burmanniaceae, Ericaceae, Gentianaceae, and Orchidaceae. Burmanniaceae and Thismiaceae should not be combined as a single family in Dioscoreales.

Phytophagous stink bugs are globally distributed and many harbor vertically inherited bacterial symbionts that are extracellular, yet little is known about how the symbiont’s genomes have evolved under this transmission strategy. Genome reduction is common in insect intracellular symbionts but limited genome sampling of the extracellular symbionts of distantly related stink bugs has precluded inferring patterns of extracellular symbiont genome evolution. To address this knowledge gap, we completely sequenced the genomes of the uncultivable bacterial symbionts of four neotropical stink bugs of the Edessa genus. Phylogenetic and comparative analyses indicated that the symbionts form a clade within the Pantoea genus and their genomes are highly reduced (∼0.8 Mb). Furthermore, genome synteny analysis and a jackknife approach for phylogenetic reconstruction, which corrected for long branch attraction artifacts, indicated that the Edessa symbionts were the result of a single symbiotic event that was distinct from the symbiosis event giving rise to Candidatus “Pantoea carbekii,” the extracellular symbiont of the invasive pentatomid stink bug, Halyomorpha halys. Metabolic functions inferred from the Edessa symbiont genomes suggests a shift in genomic composition characteristic of its lifestyle in that they retained many host-supportive functions while undergoing dramatic gene loss and establishing a stable relationship with their host insects. Given the undersampled nature of extracellular insect symbionts, this study is the first comparative analysis of these symbiont genomes from four distinct Edessa stink bug species. Finally, we propose the candidate name “Candidatus Pantoea edessiphila” for the species of these symbionts with strain designations according to their host species.

Inversions of the origin of replication (ORI) in mitochondrial genomes produce asymmetrical mutational pressures that can cause strong base composition skews. Due to skews often being overlooked, the total number of crustacean lineages that underwent ORI events remains unknown. We analysed skews, cumulative skew plots, conserved sequence motifs, and mitochondrial architecture of all 965 available crustacean mitogenomes (699 unique species). We found indications of an ORI in 159 (22.7%) species, and mapped these to 23 ORI events: 16 identified with confidence and 7 putative (13 newly proposed, and for 5 we improved the resolution). Two ORIs occurred at or above the order level: Isopoda and Copepoda. Shifts in skew plots are not a precise tool for identifying the replication mechanism. We discuss how ORIs can produce mutational bursts in mitogenomes and show how these can interfere with various types of evolutionary studies. Phylogenetic analyses were plagued by artefactual clustering, and ORI lineages exhibited longer branches, a higher number of synonymous substitutions, higher mutational saturation, and higher compositional heterogeneity. ORI events also affected codon usage and protein properties. We discuss how this may have caused erroneous interpretation of data in previous studies that did not account for skew patterns.

Echinocereus is a morphologically diverse genus that includes 64 species grouped into eight taxonomic sections based on morphological traits. In previous molecular phylogenetic analyses, the relati...

Mitogenomic trees for Bivalvia have proved problematic in the past, but several highly divergent lineages were missing from these analyses and increased representation of these groups may yet improve resolution. Here, we add seven new sequences from the Anomalodesmata and one unidentified semelid species ( Bryopa lata, Euciroa cf. queenslandica, Laternula elliptica, Laternula truncata, Lyonsia norwegica, Myadora brevis, Tropidomya abbreviata, “Abra” sp.). We show that relationships in a mitogenomic tree for the Class are improved by the addition of seven anomalodesmatans from this highly divergent clade, but are still not completely consistent with relationships recovered in studies of nuclear genes. We suggest that some anomalous relationships (for instance the non-monophyly of Bivalvia) may be partially explained by compositional heterogeneity in the mitogenome and suggest that the addition of more taxa may help resolve both this effect and possible instances of long branch attraction. We also identify several curious features about anomalodesmatan mitogenomes. For example, many protein-coding gene boundaries are poorly defined in marine bivalves, but particularly so in anomalodesmatans, primarily due to non-conserved boundary sequences. The use of transcriptomic and genomic data together enabled better definition of gene boundaries, the identification of possible pseudogenes and suggests that most genes are translated monocistronically, which contrasts with many other studies. We also identified a possible case of gene duplication of ND5 in Myadora brevis (Myochamidae) . Mitogenome size in the Anomalodesmata ranges from very small compact molecules, with the smallest for Laternula elliptica (Laternulidae) only 14,622 bp, to Bryopa lata (Clavagellidae) which is at least 31,969 bp long and may be >40,000 bp. Finally, sampled species show a high degree of sequence divergence and variable gene order, although intraspecific variation in Laternula elliptica is very low.

The story of how preColumbian civilizations developed goes hand-in-hand with the process of plant domestication by Mesoamerican inhabitants. Here, we present the almost complete sequence of a mitochondrial genome and a partial chloroplast genome from an archaeological maize sample collected at the Valley of Tehuacan, Mexico. Accelerator mass spectrometry dated the maize sample to be 5,040-5,300 years before present (95% probability). Phylogenetic analysis of the mitochondrial genome shows that the archaeological sample branches basal to the other Zea mays genomes, as expected. However, this analysis also indicates that fertile genotype NB is closely related to the archaeological maize sample and evolved before cytoplasmic male sterility genotypes (CMS-S, CMS-T, and CMS-C), thus contradicting previous phylogenetic analysis of mitochondrial genomes from maize. We show that maximum-likelihood infers a tree where CMS genotypes branch at the base of the tree when including sites that have a relative fast rate of evolution thus suggesting long-branch attraction. We also show that Bayesian analysis infer a topology where NB and the archaeological maize sample are at the base of the tree even when including faster sites. We therefore suggest that previous trees suffered from long-branch attraction. We also show that the phylogenetic analysis of the ancient chloroplast is congruent with genotype NB to be more closely related to the archaeological maize sample. As shown here, the inclusion of ancient genomes on phylogenetic trees greatly improves our understanding of the domestication process of maize, one of the most important crops worldwide.

Resolving deep divergences in the tree of life is challenging even for analyses of genome-scale phylogenetic data sets. Relationships between Basidiomycota subphyla, the rusts and allies (Pucciniomycotina), smuts and allies (Ustilaginomycotina), and mushroom-forming fungi and allies (Agaricomycotina) were found particularly recalcitrant both to traditional multigene and genome-scale phylogenetics. Here, we address basal Basidiomycota relationships using concatenated and gene tree-based analyses of various phylogenomic data sets to examine the contribution of several potential sources of bias. We evaluate the contribution of biological causes (hard polytomy, incomplete lineage sorting) versus unmodeled evolutionary processes and factors that exacerbate their effects (e.g., fast-evolving sites and long-branch taxa) to inferences of basal Basidiomycota relationships. Bayesian Markov Chain Monte Carlo and likelihood mapping analyses reject the hard polytomy with confidence. In concatenated analyses, fast-evolving sites and oversimplified models of amino acid substitution favored the grouping of smuts with mushroom-forming fungi, often leading to maximal bootstrap support in both concatenation and coalescent analyses. On the contrary, the most conserved data subsets grouped rusts and allies with mushroom-forming fungi, although this relationship proved labile, sensitive to model choice, to different data subsets and to missing data. Excluding putative long-branch taxa, genes with high proportions of missing data and/or with strong signal failed to reveal a consistent trend toward one or the other topology, suggesting that additional sources of conflict are at play. While concatenated analyses yielded strong but conflicting support, individual gene trees mostly provided poor support for any resolution of rusts, smuts, and mushroom-forming fungi, suggesting that the true Basidiomycota tree might be in a part of tree space that is difficult to access using both concatenation and gene tree-based approaches. Inference-based assessments of absolute model fit strongly reject best-fit models for the vast majority of genes, indicating a poor fit of even the most commonly used models. While this is consistent with previous assessments of site-homogenous models of amino acid evolution, this does not appear to be the sole source of confounding signal. Our analyses suggest that topologies uniting smuts with mushroom-forming fungi can arise as a result of inappropriate modeling of amino acid sites that might be prone to systematic bias. We speculate that improved models of sequence evolution could shed more light on basal splits in the Basidiomycota, which, for now, remain unresolved despite the use of whole genome data.

Maximum likelihood and maximum parsimony are two key methods for phylogenetic tree reconstruction. Under certain conditions, each of these two methods can perform more or less efficiently, resulting in unresolved or disputed phylogenies. We show that a neural network can distinguish between four-taxon alignments that were evolved under conditions susceptible to either long-branch attraction or long-branch repulsion. When likelihood and parsimony methods are discordant, the neural network can provide insight as to which tree reconstruction method is best suited to the alignment. When applied to the contentious case of Strepsiptera evolution, our method shows robust support for the current scientific view, that is, it places Strepsiptera with beetles, distant from flies.

Phylogenetic inference is of fundamental importance to evolutionary as well as other fields of biology, and molecular sequences have emerged as the primary data for this task. Although many phylogenetic methods have been developed to explicitly take into account substitution models of sequence evolution, such methods could fail due to model misspecification and insufficiency, especially in the face of heterogeneities in substitution processes across sites and among lineages. In this study, we propose to infer topologies of four-taxon trees using deep residual neural networks, a machine learning approach needing no explicit modeling of the subject system and having a record of success in solving complex non-linear inference problems. We train residual networks on simulated protein sequence data with extensive amino acid substitution heterogeneities. We show that the well-trained residual network predictors can outperform existing state-of-the-art inference methods such as the maximum likelihood method on diverse simulated test data, especially under extensive substitution heterogeneities. Reassuringly, residual network predictors generally agree with existing methods in the trees inferred from real phylogenetic data with known or widely believed topologies. We conclude that deep learning represents a powerful new approach to phylogenetic reconstruction, especially when sequences evolve via heterogeneous substitution processes. We present our best trained predictor in a freely available program named Phylogenetics by Deep Learning (PhyDL, https://gitlab.com/ztzou/phydl).

[EN] Long-branch attraction (LBA) is a well-known artifact in phylogenetic reconstruction. Sparse taxon sampling and extreme heterogeneity of evolutionary rates among lineages generate propitious situations for LBA, even defying probabilistic methods of phylogenetic inference. A clear example illustrating LBA challenges is the difficulty of reconstructing the deep gastropod phylogeny, particularly using mitochondrial (mt) genomes. Previous studies consistently obtained unorthodox phylogenetic relationships due to the LBA between the mitogenomes of patellogastropods (true limpets, represented only by Lottia digitalis), heterobranchs, and outgroup taxa. Here, we use the reconstruction of the gastropod mitogenomic phylogeny as a case exercise to test the effect of key methodological approaches proposed to counteract LBA, including the selection of slow-evolving representatives, the use of different outgroups, the application of site-heterogeneous evolutionary models, and the removal of fast-evolving sites. In this regard, we sequenced three new patellogastropod mt genomes, which displayed shorter branches than the one of Lottia as well as gene organizations more similar to that of the hypothetical gastropod ancestor. Phylogenetic analyses incorporating the mt genomes of Patella ferruginea, Patella vulgata, and Cellana radiata allowed eliminating the artificial clustering of Patellogastropoda and Heterobranchia that had prevailed in previous studies. Furthermore, the use of site-heterogeneous models with certain combinations of lineages within the outgroup allowed eliminating also the LBA between Heterobranchia and the outgroup, and recovering Apogastropoda (i.e., Caenogastropoda + Heterobranchia). Hence, for the first time, we were able to obtain a mitogenomic phylogeny of gastropods that is congruent with both morphological and nuclear datasets., The “Juan de la Cierva-Incorporación” postdoctoral fellowship IJCI- 2016-29566 to II

Phylogenetic inference artifacts can occur when sequence evolution deviates from assumptions made by the models used to analyze them. The combination of strong model assumption violations and highly heterogeneous lineage evolutionary rates can become problematic in phylogenetic inference, and lead to the well-described long-branch attraction (LBA) artifact. Here, we define an objective criterion for assessing lineage evolutionary rate heterogeneity among predefined lineages: the result of a likelihood ratio test between a model in which the lineages evolve at the same rate (homogeneous model) and a model in which different lineage rates are allowed (heterogeneous model). We implement this criterion in the algorithm Locus Specific Sequence Subsampling (LS³), aimed at reducing the effects of LBA in multi-gene datasets. For each gene, LS³ sequentially removes the fastest-evolving taxon of the ingroup and tests for lineage rate homogeneity until all lineages have uniform evolutionary rates. The sequences excluded from the homogeneously evolving taxon subset are flagged as potentially problematic. The software implementation provides the user with the possibility to remove the flagged sequences for generating a new concatenated alignment. We tested LS³ with simulations and two real datasets containing LBA artifacts: a nucleotide dataset regarding the position of Glires within mammals and an amino-acid dataset concerning the position of nematodes within bilaterians. The initially incorrect phylogenies were corrected in all cases upon removing data flagged by LS³.

Large taxa-rich genome-scale data sets are often necessary for resolving ancient phylogenetic relationships. But accurate phylogenetic inference requires that they are analyzed with realistic models that account for the heterogeneity in substitution patterns amongst the sites, genes and lineages. Two kinds of adjustments are frequently used: models that account for heterogeneity in amino acid frequencies at sites in proteins, and partitioned models that accommodate the heterogeneity in rates (branch lengths) among different proteins in different lineages (protein-wise heterotachy). Although partitioned and site-heterogeneous models are both widely used in isolation, their relative importance to the inference of correct phylogenies has not been carefully evaluated. We conducted several empirical analyses and a large set of simulations to compare the relative performances of partitioned models, site-heterogeneous models, and combined partitioned site heterogeneous models. In general, site-homogeneous models (partitioned or not) performed worse than site heterogeneous, except in simulations with extreme protein-wise heterotachy. Furthermore, simulations using empirically-derived realistic parameter settings showed a marked long-branch attraction (LBA) problem for analyses employing protein-wise partitioning even when the generating model included partitioning. This LBA problem results from a small sample bias compounded over many single protein alignments. In some cases, this problem was ameliorated by clustering similarly-evolving proteins together into larger partitions using the PartitionFinder method. Similar results were obtained under simulations with larger numbers of taxa or heterogeneity in simulating topologies over genes. For an empirical Microsporidia test data set, all but one tested site-heterogeneous models (with or without partitioning) obtain the correct Microsporidia+Fungi grouping, whereas site-homogenous models (with or without partitioning) did not. The single exception was the fully partitioned site-heterogeneous analysis that succumbed to the compounded small sample LBA bias. In general unless protein-wise heterotachy effects are extreme, it is more important to model site-heterogeneity than protein-wise heterotachy in phylogenomic analyses. Complete protein-wise partitioning should be avoided as it can lead to a serious LBA bias. In cases of extreme protein-wise heterotachy, approaches that cluster similarly-evolving proteins together and coupled with site-heterogeneous models work well for phylogenetic estimation.

Horseshoe crabs (Xiphosura) are traditionally regarded as sister group to the clade of terrestrial chelicerates (Arachnida). This hypothesis has been challenged by recent phylogenomic analyses, but the non-monophyly of Arachnida has consistently been disregarded as artifactual. We re-evaluated the placement of Xiphosura among chelicerates using the most complete phylogenetic data set to date, expanding outgroup sampling, and including data from whole genome sequencing projects. In spite of uncertainty in the placement of some arachnid clades, all analyses show Xiphosura consistently nested within Arachnida as the sister group to Ricinulei (hooded tick spiders). It is apparent that the radiation of arachnids is an old one and occurred over a brief period of time, resulting in several consecutive short internodes, and thus is a potential case for the confounding effects of incomplete lineage sorting (ILS). We simulated coalescent gene trees to explore the effects of increasing levels of ILS on the placement of horseshoe crabs. In addition, common sources of systematic error were evaluated, as well as the effects of fast-evolving partitions and the dynamics of problematic long branch orders. Our results indicated that the placement of horseshoe crabs cannot be explained by missing data, compositional biases, saturation, or ILS. Interrogation of the phylogenetic signal showed that the majority of loci favor the derived placement of Xiphosura over a monophyletic Arachnida. Our analyses support the inference that horseshoe crabs represent a group of aquatic arachnids, comparable to aquatic mites, breaking a long-standing paradigm in chelicerate evolution and altering previous interpretations of the ancestral transition to the terrestrial habitat. Future studies testing chelicerate relationships should approach the task with a sampling strategy where the monophyly of Arachnida is not held as the premise.

Statistically consistent estimation of phylogenetic trees or gene trees is possible if pairwise sequence dissimilarities can be converted to a set of distances that are proportional to the true evolutionary distances. Susko et al. (2004) reported some strikingly broad results about the forms of inconsistency in tree estimation that can arise if corrected distances are not proportional to the true distances. They showed that if the corrected distance is a concave function of the true distance, then inconsistency due to long branch attraction will occur. If these functions are convex, then two "long branch repulsion" trees will be preferred over the true tree - though these two incorrect trees are expected to be tied as the preferred true. Here we extend their results, and demonstrate the existence of a tree shape (which we refer to as a "twisted Farris-zone" tree) for which a single incorrect tree topology will be guaranteed to be preferred if the corrected distance function is convex. We also report that the standard practice of treating gaps in sequence alignments as missing data is sufficient to produce non-linear corrected distance functions if the substitution process is not independent of the insertion/deletion process. Taken together, these results imply inconsistent tree inference under mild conditions. For example, if some positions in a sequence are constrained to be free of substitutions and insertion/deletion events while the remaining sites evolve with independent substitutions and insertion/deletion events, then the distances obtained by treating gaps as missing data can support an incorrect tree topology even given an unlimited amount of data.

It has recently been concluded that phylogenomic data from 310 nuclear genes support the clade of (Amborellales, Nymphaeales) as sister to the remaining angiosperms and that shortcut coalescent phylogenetic methods outperformed concatenation for these data. We falsify both of those conclusions here by demonstrating that discrepant results between the coalescent and concatenation analyses are primarily caused by the coalescent methods applied (MP-EST and STAR) not being robust to the highly divergent and often mis-rooted gene trees that were used. This result reinforces the expectation that low amounts of phylogenetic signal and methodological artifacts in gene-tree reconstruction can be more problematic for shortcut coalescent methods than is the assumption of a single hierarchy for all genes by concatenation methods when these approaches are applied to ancient divergences in empirical studies. We also demonstrate that a third coalescent method, ASTRAL, is more robust to mis-rooted gene trees than MP-EST or STAR, and that both Observed Variability (OV) and Tree Independent Generation of Evolutionary Rates (TIGER), which are two character subsampling procedures, are biased in favor of characters with highly asymmetrical distributions of character states when applied to this dataset. We conclude that enthusiastic application of novel tools is not a substitute for rigorous application of first principles, and that trending methods (e.g., shortcut coalescent methods applied to ancient divergences, tree-independent character subsampling), may be novel sources of previously under-appreciated, systematic errors.

For high-level molecular phylogenies, a comprehensive sampling design is a key factor for not only improving inferential accuracy, but also for maximizing the explanatory power of the resulting phylogeny. Two standing problems in molecular phylogenies are the unstable placements of some deep and long branches, and the phylogenetic relationships shown by robust supported clades conflict with recognized knowledge. Empirical and theoretical studies suggest that increasing taxon sampling is expected to ameliorate, if not resolve, both problems; however, sometimes neither the current taxonomic system nor the established phylogeny can provide sufficient information to guide additional sampling design. We examined the phylogeny of the spider family Linyphiidae, and selected ingroup species based on epigynal morphology, which can be reconstructed in a phylogenetic context. Our analyses resulted in seven robustly supported clades within linyphiids. The placements of four deep and long branches are sensitive to variations in both outgroup and ingroup sampling, suggesting the possibility of long branch attraction artifacts. Results of ancestral state reconstruction indicate that successive state transformations of the epigynal plate are associated with early cladogenetic events in linyphiid diversification. Representatives of different subfamilies were mixed together within well supported clades and examination revealed that their defining characters, as per traditional taxonomy, are homoplastic. Furthermore, our results demonstrated that increasing taxon sampling produced a more informative framework, which in turn helps to study character evolution and interpret the relationships among linyphiid lineages. Additional defining characters are needed to revise the linyphiid taxonomic system based on our phylogenetic hypothesis.

With advances in sequencing technologies, there are now massive amounts of genomic data from across all life, leading to the possibility that a robust Tree of Life can be constructed. However, "gene tree heterogeneity", which is when different genomic regions can evolve differently, is a common phenomenon in multi-locus datasets, and reduces the accuracy of standard methods for species tree estimation that do not take this heterogeneity into account. New methods have been developed for species tree estimation that specifically address gene tree heterogeneity, and that have been proven to converge to the true species tree when the number of loci and number of sites per locus both increase (i.e., the methods are said to be "statistically consistent"). Yet, little is known about the biologically realistic condition where the number of sites per locus is bounded. We show that when the sequence length of each locus is bounded (by any arbitrarily chosen value), the most common approaches to species tree estimation that take heterogeneity into account (i.e., traditional fully partitioned concatenated maximum likelihood and newer approaches, called summary methods, that estimate the species tree by combining gene trees) are not statistically consistent, even when the heterogeneity is extremely constrained. The main challenge is the presence of conditions such as long branch attraction that create biased tree estimation when the number of sites is restricted. Hence, our study uncovers a fundamental challenge to species tree estimation using both traditional and new methods., Comment: Submitted

The placement of angiosperms and Gnetales in seed plant phylogeny remains one of the most enigmatic problems in plant evolution, with morphological analyses (which have usually included fossils) and molecular analyses pointing to very distinct topologies. Almost all morphology-based phylogenies group angiosperms with Gnetales and certain extinct seed plant lineages, while most molecular phylogenies link Gnetales with conifers. In this study, we investigate the phylogenetic signal present in published seed plant morphological datasets. We use parsimony, Bayesian inference, and maximum likelihood approaches, combined with a number of experiments with the data, to address the morphological-molecular conflict. First, we ask whether the lack of association of Gnetales with conifers in morphological analyses is due to an absence of signal or to the presence of competing signals, and second, we compare the performance of parsimony and model based approaches with morphological datasets. Our results imply that the grouping of Gnetales and angiosperms is largely the result of long branch attraction, consistent across a range of methodological approaches. Thus, there is a signal for the grouping of Gnetales with conifers in morphological matrices, but it was swamped by convergence between angiosperms and Gnetales, both situated on long branches. However, this effect becomes weaker in more recent analyses, as a result of addition and critical reassessment of characters. Even when a clade including angiosperms and Gnetales is still weakly supported by parsimony, model-based approaches favor a clade of Gnetales and conifers, presumably because they are more resistant to long branch attraction. Inclusion of fossil taxa weakens rather than strengthens support for a relationship of angiosperms and Gnetales. Our analyses finally reconcile morphology with molecules in favoring a relationship of Gnetales to conifers, and show that morphology may therefore be useful in reconstructing other aspects of the phylogenetic history of the seed plants.

International audience; Phylogenetic analyses of conserved core genes have disentangled most of the ancient relationships in Archaea. However, some groups remain debated, like the DPANN, a deep-branching super-phylum composed of nanosized archaea with reduced genomes. Among these, the Nanohaloarchaea require high-salt concentrations for growth. Their discovery in 2012 was significant because they represent, together with Halobacteria (a Class belonging to Euryarchaeota), the only two described lineages of extreme halophilic archaea. The phylogenetic position of Nanohaloarchaea is highly debated, being alternatively proposed as the sister-lineage of Halobacteria or a member of the DPANN super-phylum. Pinpointing the phylogenetic position of extreme halophilic archaea is important to improve our knowledge of the deep evolutionary history of Archaea and the molecular adaptive processes and evolutionary paths that allowed their emergence. Using comparative genomic approaches, we identified 258 markers carrying a reliable phylogenetic signal. By combining strategies limiting the impact of biases on phylogenetic inference, we showed that Nanohaloarchaea and Halobacteria represent two independent lines that derived from two distinct but related methanogen Class II lineages. This implies that adaptation to high salinity emerged twice independently in Archaea and indicates that emergence of Nanohaloarchaea within DPANN in previous studies is likely the consequence of a tree reconstruction artifact, challenging the existence of this super-phylum.

Previous work on the star-tree paradox has shown that Bayesian methods suffer from a long branch attraction bias. That work is extended to settings involving more taxa and partially resolved trees. The long branch attraction bias is confirmed to arise more broadly and an additional source of bias is found. A by-product of the analysis is methods that correct for biases toward particular topologies. The corrections can be easily calculated using existing Bayesian software. Posterior support for a set of two or more trees can thus be supplemented with corrected versions to cross-check or replace results. Simulations show the corrections to be highly effective.

© 2014 by Annual Reviews. All rights reserved. Originating most likely in the early Cretaceous, ants have diversified to become the world's most successful eusocial insects, occupying most terrestrial ecosystems and acquiring a global ecological footprint. Recent advances in our understanding of ant evolutionary history have been propelled by the use of molecular phylogenetic methods, in conjunction with a rich (and still growing) fossil record. Most extant ants belong to the formicoid clade, which contains ∼90% of described species and has produced the most socially advanced and dominant forms. The remaining ants are old lineages of predominantly cryptobiotic species whose relationships to one another and to the formicoids remain unclear. Rooting the ant tree is challenging because of (a) a long branch separating ants from their nearest outgroup, and (b) heterogeneity in evolutionary rates and base composition among ant lineages. These factors will need to be given careful consideration in future phylogenomic studies of ants.

We review the morphological and molecular evidence that Mayden & Chen recently used to infer that the developmentally truncated fish genus Paedocypris is not a member of the teleost order Cypriniformes or carp-like fishes, but is ‘the basal sister group to all Cypriniformes’. This hypothesis contradicts several previous studies that used molecular sequence data or morphological characters. A review of the morphological characters that Mayden & Chen discussed and mapped onto their ‘simplified tree’ shows that these, analysed alone, rather support a close relationship of the cyprinids Sundadanio, Danionella, and Paedocypris. We also present four additional analyses of morphological data, which all contradict Mayden & Chen's result. Despite its highly reductive skeleton, posing a serious problem when analysing its phylogenetic position with skeletal characters, the presence in Paedocypris of the basioccipital masticatory plate is compelling evidence that it is a member of the Cyprinoidei (Cyprinidae plus Psilorhynchidae). Our reanalysis and exploration of their molecular sequence data shows that only a single gene, EGR3, of the six nuclear genes analysed by Mayden & Chen, is responsible for the position of Paedocypris as ‘the basal sister group to all Cypriniformes’. Three independent methods to visualize and analyse phylogenetic signal and conflict of data sets (phylogenetic networks, splits analysis methods or SAMS, and site-wise likelihood analyses) reveal a high level of character conflict and noise in Mayden & Chen's data set. The ‘basal’ position of Paedocypris seems to be the outcome of the interplay of two long-branch effects. We apply the same analytical methods to the data set from Ruber et al.'s molecular analysis of the phylogenetic position of Paedocypris and discuss our findings. We conclude that none of the molecular data sets compiled to date can establish the phylogenetic position of Paedocypris with confidence. Morphological data suggest that Paedocypris and Danionella are sister genera, and that their closest relative is Sundadanio, although the position of these three miniatures among cyprinoids is still unclear. © 2014 The Linnean Society of London

Diurodrilidae is a taxon of Lophotrochozoa comprising about six, exclusively interstitial species, which are up to 500μm long and dorsoventrally flattened. Traditionally, Diurodrilidae had been regarded as an annelid family. However, recently Diurodrilidae had been excluded from Annelida and been placed in closer relationship to platyzoan taxa based on both morphological and nuclear rRNA data. Since both, Diurodrilidae and platyzoan taxa, exhibit long branches in the molecular analyses, the close relationship might be due to a long branch attraction artifact. The annelid taxon Myzostomida had been trapped in a similar long branch attraction artifact with platyzoan taxa using nuclear rRNA data, but determination of the nearly complete mitochondrial genome of myzostomids revealed their annelid affinity. Therefore, we determined the nearly complete mitochondrial genome of Diurodrilus subterraneus as well as new nuclear rRNA data for D. subterraneus and some platyzoan taxa. All our analyses of nuclear rRNA and mitochondrial sequence and gene order data presented herein clearly place Diurodrilidae within Annelida and with strong nodal support values in some analyses. Therefore, the previously suggested exclusion of Diurodrilidae from Annelida and its close relationship with platyzoan taxa can be attributed to a long branch artifact. Morphological data do not unambiguously support a platyzoan affinity of Diurodrilidae, but instead would also be in line with a progenetic origin of Diurodrilidae within Annelida.

In a recent contribution to PLoS Computational Biology, Cannarozzi, Schneider, and Gonnet published evidence that rodents form an outgroup to human and dog [1], in disagreement with several recent studies suggesting that the dog is an outgroup to the primate–rodent clade [2,3]. The authors' arguments rest on a variety of analyses of human, mouse, and dog genes, using opossum to root the phylogeny. Here I argue that despite the large number of characters used in this study, their results may well be erroneous. I then provide new and, I believe, conclusive evidence in favour of the current consensus phylogeny, and I briefly review other recent studies that support this conclusion. The problem of determining the evolutionary relationship between all extant mammals has a long history. Traditionally, morphological features were used to group “like” mammals together in a tree, purportedly reflecting their phylogeny. More recently, molecular data have generally confirmed these inferences, but have also led to surprising revisions. While sequence analysis is more objective than morphology, it nevertheless emerged that it has its own set of issues, and some phylogenies remain contentious. In [1], Cannarozzi et al. suggested that this contention extends to the phylogeny of human, mouse, and dog, and inferred a phylogeny of these species that disagrees with a recently emerging consensus. Here I challenge their findings, providing new evidence in support of the consensus phylogeny, and suggest that their results may have been biased by long branch attraction (LBA), a known issue in molecular phylogenetic inference. It is well-known that phylogenetic inferences can be biased, and may be inaccurate even with strong bootstrap or posterior support. Felsenstein showed that in parsimony analyses, long branches in the phylogeny tend to attract one another [4]. In contrast to what the authors claim, maximum likelihood methods, although less vulnerable, are similarly affected by LBA [5], particularly when small numbers of taxa are used [6]. This methodological bias has led to various erroneous inferences, such as the now-discredited claim that “the guinea pig is not a rodent” [7,8]. Perhaps counterintuitively, the effect of LBA does not diminish with increasing amounts of sequence data. To quote from a review, “spurious conclusions are often derived from an over-credibility of enormous numbers of nucleotide or amino acid characters (e.g., complete genomes) when combined with poor taxon sampling” [9]. The recently emerging consensus on mammalian phylogeny based on molecular data is surprisingly different from the traditional, morphological phylogeny [2,3]. It proposes four mammalian cohorts, including the Laurasiatheria (of which the dog lineage is part), which separated from the Euarchontoglires about 85–95 million years ago (Mya) [10]. The subsequent speciation separating the Euarchontoglires into Glires (including rodents) and Euarchonta (which includes primates) occurred roughly 80 Mya. The difference is small compared with the total branch length to opossum (180 + 90 My), so that a relatively small bias would suffice to bring about a topology change. As the mouse genome sequence has been evolving fast relative to those of human and dog [11], its branch is expected to be affected by LBA to the opossum branch, which would result in the reported grouping. These considerations throw some doubt on both the parsimony and maximum likelihood analyses. What about the genome rearrangement argument? After all, genome rearrangements are large-scale but relatively infrequent events, so that the parsimony approximation might be justified. However, the opossum genome had not yet been assembled, and the authors had to resort to chicken, which diverged ∼310 Mya from the mammalian lineage, considerably earlier than the opossum did. Moreover, there is strong evidence for hotspots of breakage [12] and breakpoint reuse [13], discounting the “random breakage” model. The use of (nuclear) gene orderings to analyze rearrangements further exacerbates these issues, as it affords little power to resolve breakpoints and artificially increases inhomogeneities in breakage rates, because of large and highly variable intergenic distances. For these reasons, the parsimony approximation may well be invalid, which makes LBA a concern for the genome rearrangement analysis, too. I thus considered whether the reported tree might be incorrect. To investigate the issue, I used a simple (and, to my knowledge, novel) summary statistic based on the distribution of transposable elements (TEs) in pairwise alignments, which does not require an outgroup genome to root the phylogeny. If a family of TEs is specific to lineage x when compared with y, each occurrence in x is expected to be located opposite a gap in a whole-genome alignment of species x to y. In contrast, if the family is ancestral to x and y, a proportion of TEs will have survived in both species and will align. To quantify the evidence for these alternatives, I defined a statistic A(y|x) (for “ancestralness”) as the proportion of nucleotides from a particular TE family in species x that is aligned to a secondary species y. This statistic is near-zero if a family of TEs is specific to x, and non-zero if it is ancestral to the species split. For an outgroup x and a particular family of TEs, the statistics A(y|x) are thus expected to be consistent across ingroup species y (either zero, or non-zero, for all). In contrast, for an ingroup species, some TE families may be ancestral with respect to another ingroup, but lineage-specific when compared with the outgroup. Provided such TE families exist, this would then determine the topology of the phylogeny. The results (Figure 1 and Table 1) show clear support for the rodent–primate grouping. For example, the MLT2B2 long terminal repeat element is clearly ancestral in the human-to-mouse and mouse-to-human comparisons (A > 0.20), but is highly lineage-specific in the other comparisons, each of which include the dog (A < 0.03 for all). This pattern can be explained if dog is assumed to be an outgroup to both human and mouse, and that the element has been active primarily between the two speciation events. The same pattern was observed for several other TE families (MLT1A0, MLT2B1, L1MA9, L1MB1, L1MC1, MER31A, MER21B, MER34), while no examples supporting alternative groupings were found. Unlike analyses based on nucleotide characters, TE-based studies are not expected to suffer from LBA, because the size of TEs allows for reliable homology assignments (if well-anchored alignments are used), and the marked differences between the TE insertion and small deletion processes means that back mutations are rare. It thus appears that the dog lineage is basal to the primate and rodent lineages. Figure 1 Evidence for the ((Human, Mouse), Dog) Phylogeny Table 1 Ancestralness of TE Families in the Six Pairwise Comparisons between Human, Dog, and Mouse Numerous recent studies support this conclusion. When many taxa are analyzed simultaneously, the dog consistently appears as an outgroup to human and mouse, when using either nuclear or mitochondrial DNA [2,3,9,14–16]. Studies of rare genomic changes (which are less vulnerable to LBA) consistently support this grouping. For example, by rooting the phylogeny using the consensus sequence of TEs, the evolutionary distance between the speciation events was estimated to be 0.024 substitutions per site [11]. In another study, two of the TE families found here, MLT1A0 and L1MA9, were identified as clear examples supporting the rodent–primate grouping [17], and a recent analysis of several single TE insertions provides additional support [18], as does a method that uses multiple alignments of TEs to infer phylogenies in very similar ways to ours [19]. Rare indels at homologous positions in otherwise well-conserved protein-coding genes also support this phylogeny [20]. Finally, a large cluster of PRAME genes that is absent in chicken and dog, but present in homologous locations in human and mouse, again support the same grouping [21]. Taken together with the possible influence of LBA on the analysis of Cannarozzi et al. [1], it appears unjustified to continue to consider the phylogeny of primates, rodents, and canines as contentious.

Long Branch Attraction (LBA) is a well-known artifact in phylogenetic reconstruction when dealing with branch length heterogeneity. Here we show another phenomenon, Short Branch Attraction (SBA), which occurs when BLAST searches, a phenetic analysis, are used as a surrogate method for phylogenetic analysis. This error also results from branch length heterogeneity, but this time it is the short branches that are attracting. The SBA artifact is reciprocal and can be returned 100% of the time when multiple branches differ in length by a factor of more than two. SBA is an intended feature of BLAST searches, but becomes an issue, when top scoring BLAST hit analyses are used to infer Horizontal Gene Transfers (HGTs), assign taxonomic category with environmental sequence data in phylotyping, or gather homologous sequences for building gene families. SBA can lead researchers to believe that there has been a HGT event when only vertical descent has occurred, cause slowly evolving taxa to be over-represented and quickly evolving taxa to be under-represented in phylotyping, or systematically exclude quickly evolving taxa from analyses. SBA also contributes to the changing results of top scoring BLAST hit analyses as the database grows, because more slowly evolving taxa, or short branches, are added over time, introducing more potential for SBA. SBA can be detected by examining reciprocal best BLAST hits among a larger group of taxa, including the known closest phylogenetic neighbors. Therefore, one should look for this phenomenon when conducting best BLAST hit analyses as a surrogate method to identify HGTs, in phylotyping, or when using BLAST to gather homologous sequences.

The order Poales comprises a substantial portion of plant life (7% of all angiosperms and 33% of monocots) and includes taxa of enormous economic and ecological significance. Molecular and morphological studies over the past two decades, however, leave uncertain many relationships within Poales and among allied commelinid orders. Here we present the results of an initial project by the Monocot AToL (Angiosperm Tree of Life) team on phylogeny and evolution in Poales, using sequence data for 81 plastid genes (exceeding 101 aligned kb) from 83 species of angiosperms. We recovered highly concordant relationships using maximum likelihood (ML) and maximum parsimony (MP), with 98.2% mean ML bootstrap support across monocots. For the first time, ML resolves ties among Poales and other commelinid orders with moderate to strong support. Analyses provide strong support for Bromeliaceae being sister to the rest of Poales; Typhaceae, Rapateaceae, and cyperids (sedges, rushes, and their allies) emerge next alo...

Recent phylogenetic analyses of molecular data have supported different hypotheses of relationships among Cornales, Ericales, and core asterids. Such hypotheses have implications for the evolution of important morphological and embryological features of asterids. In this study we generated plastid genome-scale data of Davidia (Cornales) and Franklinia (Ericales) and combined them with published sequence data of eudicots. Our maximum parsimony, maximum likelihood, and Bayesian analyses generated strongly supported and congruent phylogenetic relationships among the three major lineages of the asterids. Cornales diverges first in asterids; Ericales is sister to the core asterids. Adding two more taxa helps mitigate long branch attraction in parsimony analyses. Sampling 26–28 plastid protein-coding genes may provide satisfactory resolution and support for relationships of eudicots including basal lineages of asterids.

Noisy data, especially in combination with misalignment and model misspecification can have an adverse effect on phylogeny reconstruction; however, effective methods to identify such data are few. One particularly important class of noisy data is saturated positions. To avoid potential errors related to saturation in phylogenomic analyses, we present an automated procedure involving the step-wise removal of the most variable positions in a given data set coupled with a stopping criterion derived from correlation analyses of pairwise ML distances calculated from the deleted (saturated) and the remaining (conserved) subsets of the alignment. Through a comparison with existing methods, we demonstrate both the effectiveness of our proposed procedure for identifying noisy data and the effect of the removal of such data using a well-publicized case study involving placental mammals. At the least, our procedure will identify data sets requiring greater data exploration, and we recommend its use to investigate the effect on phylogenetic analyses of removing subsets of variable positions exhibiting weak or no correlation to the rest of the alignment. However, we would argue that this procedure, by identifying and removing noisy data, facilitates the construction of more accurate phylogenies by, for example, ameliorating potential long-branch attraction artefacts.

 Premise of the Study: The monilophytes (ferns and relatives) — the third largest group of land plants — exhibit a diverse array of vegetative and reproductive morphologies. Investigations into their early ecological and life-history diversifi cation require accurate, well-corroborated phylogenetic estimates. We examined the utility of a large plastid-based data set in inferring backbone relationships for monilophytes.  Methods: We recovered 17 plastid genes for exemplar taxa using published and new primers. We compared results from maximum-likelihood and parsimony analyses, assessed the effects of removing rapidly evolving characters, and examined the extent to which our data corroborate or contradict the results of other studies, or resolve current ambiguities.  Key Results: Considering multifamily clades, we found bootstrap support comparable to or better than that in published studies that used fewer genes from fewer or more taxa. We fiestablish fi lmy ferns (Hymenophyllales) as the sister group of all leptosporangiates except Osmundaceae, resolving the second deepest split in leptosporangiate-fern phylogeny. A clade comprising Ophioglossaceae and Psilotaceae is currently accepted as the sister group of other monilophytes, but we recover Equisetum in this position. We also recover marattioid and leptosporangiate ferns as sister groups. Maximum-likelihood rate-class estimates are somewhat skewed when a long-branch lineage ( Selaginella ) is included, negatively affecting bootstrap support for early branches.  Conclusions: Our fi ndings support the utility of this gene set in corroborating relationships found in previous studies, improving support, and resolving uncertainties in monilophyte phylogeny. Despite these advances, our results also underline the need for continued work on resolving the very earliest splits in monilophyte phylogeny.

The phylogenetic position of Gnetales is one of the most contentious issues in the seed plant systematics. To elucidate the Gnetales position, an improved amino acid substitution matrix was estimated based on 64 chloroplast (cp) genomes and was applied to cp genome data including all three lineages of Gnetales in maximum likelihood analyses of proteins. Although the initial analysis strongly supported the sister relation of Gnetales with Cryptomeria (Cupressophyta or non-Pinaceae conifers) (the "Gnecup" hypothesis), the support seems to be caused by a long-branch attraction (LBA) artifact. Indeed, by removing fastest evolving proteins that are most likely associated with the LBA, the support drastically declined. Furthermore, another analysis of partial genome data with dense taxon sampling of conifers showed that, in psbC, rpl2, and rps7 proteins, there are many parallel amino acid substitutions between the lineages leading to Gnetales and to Cryptomeria, and by further excluding these three genes, the sister relation of Gnetales with Pinaceae (the "Gnepine" hypothesis) became supported. Overall, our analyses indicate that the LBA and parallel substitutions cause a seriously biased inference of phylogenetic position of Gnetales with the cp genome data.

Acariformes (one of the two main lineages of Acari) represent an exceptionally diverse group of Arachnida. We performed first comprehensive phylogenetic analysis of Acariformes using sequence data from the nuclear small subunit rRNA gene (18S rDNA) and the mitochondrial cytochrome c oxidase subunit I (COI, amino acids). Our analyses confirm the monophyly of Acariformes and recognize two orders within Acariformes: Sarcoptiformes, consisting of Endeostigmata and Oribatida + Astigmata, and Trombidiformes. The data revealed the origin of Astigmata within Oribatida with the desmonomatan superfamily Crotonioidea as the source of astigmatan radiation and the sexual family Hermanniidae as the sister group, which generally supports previous morphological hypotheses. These results were found despite the strong conflict between long-branch attraction (LBA) artifacts and phylogenetic signal. It is likely that the conflict resulted from differences in the substitution rates among acariform lineages, especially comparing slowly evolving Oribatida with rapidly evolving Astigmata. The use of likelihood methods considered more resistant to LBA only slightly decreased the chance of falling into the LBA trap; the probability of recovering the origin of Astigmata within Desmonomata differs only by about 10% from that of having the long branched Astigmata and Trombidiformes either connected directly or shifted to deep parts of the tree due to outgroup attraction. Molecular dating using the rate-smoothing method PATHd8 shows that Acariformes originated c . 435 MYA and were probably among the earliest arthropods invading terrestrial habitats in late Silurian or the Lower Devonian, when the first vascular plants are thought to have arisen. Our analyses did not support the monophyly of Acari because we recovered clades Acariformes–Solifugae and Parasitiformes–Pseudoscorpionida. However, a formal revision of arachnid classification that would reflect these results must await future analyses.

Ecdysozoa is the recently recognized clade of molting animals that comprises the vast majority of extant animal species and the most important invertebrate model organisms--the fruit fly and the nematode worm. Evolutionary relationships within the ecdysozoans remain, however, unresolved, impairing the correct interpretation of comparative genomic studies. In particular, the affinities of the three Panarthropoda phyla (Arthropoda, Onychophora, and Tardigrada) and the position of Myriapoda within Arthropoda (Mandibulata vs. Myriochelata hypothesis) are among the most contentious issues in animal phylogenetics. To elucidate these relationships, we have determined and analyzed complete or nearly complete mitochondrial genome sequences of two Tardigrada, Hypsibius dujardini and Thulinia sp. (the first genomes to date for this phylum); one Priapulida, Halicryptus spinulosus; and two Onychophora, Peripatoides sp. and Epiperipatus biolleyi; and a partial mitochondrial genome sequence of the Onychophora Euperipatoides kanagrensis. Tardigrada mitochondrial genomes resemble those of the arthropods in term of the gene order and strand asymmetry, whereas Onychophora genomes are characterized by numerous gene order rearrangements and strand asymmetry variations. In addition, Onychophora genomes are extremely enriched in A and T nucleotides, whereas Priapulida and Tardigrada are more balanced. Phylogenetic analyses based on concatenated amino acid coding sequences support a monophyletic origin of the Ecdysozoa and the position of Priapulida as the sister group of a monophyletic Panarthropoda (Tardigrada plus Onychophora plus Arthropoda). The position of Tardigrada is more problematic, most likely because of long branch attraction (LBA). However, experiments designed to reduce LBA suggest that the most likely placement of Tardigrada is as a sister group of Onychophora. The same analyses also recover monophyly of traditionally recognized arthropod lineages such as Arachnida and of the highly debated clade Mandibulata.

When cyanobacteria originated and diversified, and what their ancient traits were, remain critical unresolved problems. Here, we used a phylogenomic approach to construct a well-resolved 'core' cyanobacterial tree. The branching positions of four lineages (Thermosynechococcus elongatus, Synechococcus elongatus, Synechococcus PCC 7335 and Acaryochloris marina) were problematic, probably due to long branch attraction artifacts. A consensus genomic tree was used to study trait evolution using ancestral state reconstruction (ASR). The early cyanobacteria were probably unicellular, freshwater, had small cell diameters, and lacked the traits to form thick microbial mats. Relaxed molecular clock analyses suggested that early cyanobacterial lineages were restricted to freshwater ecosystems until at least 2.4 Ga, before diversifying into coastal brackish and marine environments. The resultant increases in niche space and nutrient availability, and consequent sedimentation of organic carbon into the deep oceans, would have generated large pulses of oxygen into the biosphere, possibly explaining why oxygen rose so rapidly. Rapid atmospheric oxidation could have destroyed the methane-driven greenhouse with simultaneous drawdown in pCO(2), precipitating 'Snowball Earth' conditions. The traits associated with the formation of thick, laminated microbial mats (large cell diameters, filamentous growth, sheaths, motility and nitrogen fixation) were not seen until after diversification of the LPP, SPM and PNT clades, after 2.32 Ga. The appearance of these traits overlaps with a global carbon isotopic excursion between 2.2 and 2.1 Ga. Thus, a massive re-ordering of biogeochemical cycles caused by the appearance of complex laminated microbial communities in marine environments may have caused this excursion. Finally, we show that ASR may provide an explanation for why cyanobacterial microfossils have not been observed until after 2.0 Ga, and make suggestions for how future paleobiological searches for early cyanobacteria might proceed. In summary, key evolutionary events in the microbial world may have triggered some of the key geologic upheavals on the Paleoproterozoic Earth.

The protistan phylum Apicomplexa contains many important pathogens and is the subject of intense genome sequencing efforts. Based upon the genome sequences from seven apicomplexan species and a ciliate outgroup, we identified 268 single-copy genes suitable for phylogenetic inference. Both concatenation and consensus approaches inferred the same species tree topology. This topology is consistent with most prior conceptions of apicomplexan evolution based upon ultrastructural and developmental characters, that is, the piroplasm genera Theileria and Babesia form the sister group to the Plasmodium species, the coccidian genera Eimeria and Toxoplasma are monophyletic and are the sister group to the Plasmodium species and piroplasm genera, and Cryptosporidium forms the sister group to the above mentioned with the ciliate Tetrahymena as the outgroup. The level of incongruence among gene trees appears to be high at first glance; only 19% of the genes support the species tree, and a total of 48 different gene-tree topologies are observed. Detailed investigations suggest that the low signal-to-noise ratio in many genes may be the main source of incongruence. The probability of being consistent with the species tree increases as a function of the minimum bootstrap support observed at tree nodes for a given gene tree. Moreover, gene sequences that generate high bootstrap support are robust to the changes in alignment parameters or phylogenetic method used. However, caution should be taken in that some genes can infer a “wrong” tree with strong support because of paralogy, model violations, or other causes. The importance of examining multiple, unlinked genes that possess a strong phylogenetic signal cannot be overstated.

Despite the prior use of approximately 9000 bp, deep-level relationships within the angiosperm clade, Saxifragales remain enigmatic, due to an ancient, rapid radiation (89.5 to 110 Ma based on the fossil record). To resolve these deep relationships, we constructed several new data sets: (1) 16 genes representing the three genomic compartments within plant cells (2 nuclear, 10 plastid, 4 mitochondrial; aligned, analyzed length = 21,460 bp) for 28 taxa; (2) the entire plastid inverted repeat (IR; 26,625 bp) for 17 taxa; (3) "total evidence" (50,845 bp) for both 17 and 28 taxa (the latter missing the IR). Bayesian and ML methods yielded identical topologies across partitions with most clades receiving high posterior probability (pp = 1.0) and bootstrap (95% to 100%) values, suggesting that with sufficient data, rapid radiations can be resolved. In contrast, parsimony analyses of different partitions yielded conflicting topologies, particularly with respect to the placement of Paeoniaceae, a clade characterized by a long branch. In agreement with published simulations, the addition of characters increased bootstrap support for the putatively erroneous placement of Paeoniaceae. Although having far fewer parsimony-informative sites, slowly evolving plastid genes provided higher resolution and support for deep-level relationships than rapidly evolving plastid genes, yielding a topology close to the Bayesian and ML total evidence tree. The plastid IR region may be an ideal source of slowly evolving genes for resolution of deep-level angiosperm divergences that date to 90 My or more. Rapidly evolving genes provided support for tip relationships not recovered with slowly evolving genes, indicating some complementarity. Age estimates using penalized likelihood with and without age constraints for the 28-taxon, total evidence data set are comparable to fossil dates, whereas estimates based on the 17-taxon data are much older than implied by the fossil record. Hence, sufficient taxon density, and not simply numerous base pairs, is important in reliably estimating ages. Age estimates indicate that the early diversification of Saxifragales occurred rapidly, over a time span as short as 6 million years. Between 25,000 and 50,000 bp were needed to resolve this radiation with high support values. Extrapolating from Saxifragales, a similar number of base pairs may be needed to resolve the many other deep-level radiations of comparable age in angiosperms.

The Siphonocladales are tropical to warm-temperate, marine green macro-algae characterized by a wide variety of thallus morphologies, ranging from branched filaments to pseudo-parenchymatous plants. Phylogenetic analyses of partial large subunit (LSU) rDNA sequences sampled from 166 isolates revealed nine well-supported siphonocladalean clades. Analyses of a concatenated dataset of small subunit (SSU) and partial LSU rDNA sequences greatly clarified the phylogeny of the Siphonocladales. However, the position of the root of the Siphonocladales could not be determined unambiguously, as outgroup rooting and molecular clock rooting resulted in a different root placement. Different phylogenetic methods (likelihood, parsimony and distance) yielded similar tree topologies with comparable internal node resolution. Likewise, analyses under more realistic models of sequence evolution, taking into account differences in evolution between stem and loop regions of rRNA, did not differ markedly from analyses using standard four-state models. The molecular phylogeny revealed that all siphonocladalean architectures may be derived from a single Cladophora-like ancestor. Parallel and convergent evolution of various morphological characters (including those traditionally employed to circumscribe the families and genera) have occurred in the Siphonocladales. Consequently, incongruence with traditional classifications, including non-monophyly in all families and most genera, was shown.