Bayesian and likelihood phylogenetic reconstructions of morphological traits are not discordant when taking uncertainty into consideration: a comment on Puttick et al

Performance of tree-building methods using a morphological dataset and a well-supported Hexapoda phylogeny

Recently, many studies have addressed the performance of phylogenetic tree-building methods (maximum parsimony, maximum likelihood, and Bayesian inference), focusing primarily on simulated data. However, for discrete morphological data, there is no consensus yet on which methods recover the phylogeny with better performance. To address this lack of consensus, we investigate the performance of different methods using an empirical dataset for hexapods as a model. As an empirical test of performance, we applied normalized indices to effectively measure accuracy (normalized Robinson-Foulds metric, nRF) and precision, which are measured via resolution, one minus Colless' consensus fork index (1-CFI). Additionally, to further explore phylogenetic accuracy and support measures, we calculated other statistics, such as the true positive rate (statistical power) and the false positive rate (type I error), and constructed receiver operating characteristic plots to visualize the relationship between these statistics. We applied the normalized indices to the reconstructed trees from the reanalyses of an empirical discrete morphological dataset from extant Hexapoda using a well-supported phylogenomic tree as a reference. Maximum likelihood and Bayesian inference applying the k-state Markov (Mk) model (without or with a discrete gamma distribution) performed better, showing higher precision (resolution). Additionally, our results suggest that most available tree topology tests are reliable estimators of the performance measures applied in this study. Thus, we suggest that likelihood-based methods and tree topology tests should be used more often in phylogenetic tree studies based on discrete morphological characters. Our study provides a fair indication that morphological datasets have robust phylogenetic signal.

Total evidence phylogeny of platyrrhine primates and a comparison of undated and tip-dating approaches

There have been multiple published phylogenetic analyses of platyrrhine primates (New World monkeys) using both morphological and molecular data, but relatively few that have integrated both types of data into a total evidence approach. Here, we present phylogenetic analyses of recent and fossil platyrrhines, based on a total evidence data set of 418 morphological characters and 10.2 kilobases of DNA sequence data from 17 nuclear genes taken from previous studies, using undated and tip-dating approaches in a Bayesian framework. We compare the results of these analyses with molecular scaffold analyses using maximum parsimony and Bayesian approaches, and we use a formal information theoretic approach to identify unstable taxa. After a posteriori pruning of unstable taxa, the undated and tip-dating topologies appear congruent with recent molecular analyses and support largely similar relationships, with strong support for Stirtonia as a stem alouattine, Neosaimiri as a stem saimirine, Cebupithecia as a stem pitheciine, and Lagonimico as a stem callitrichid. Both analyses find three Greater Antillean subfossil platyrrhines (Xenothrix, Antillothrix, and Paralouatta) to form a clade that is related to Callicebus, congruent with a single dispersal event by the ancestor of this clade to the Greater Antilles. They also suggest that the fossil Proteropithecia may not be closely related to pitheciines, and that all known platyrrhines older than the Middle Miocene are stem taxa. Notably, the undated analysis found the Early Miocene Panamacebus (currently recognized as the oldest known cebid) to be unstable, and the tip-dating analysis placed it outside crown Platyrrhini. Our tip-dating analysis supports a late Oligocene or earliest Miocene (20.8-27.0 Ma) age for crown Platyrrhini, congruent with recent molecular clock analyses.

Evaluating the Performance of Probabilistic Algorithms for Phylogenetic Analysis of Big Morphological Datasets: A Simulation Study

Reconstructing the tree of life is an essential task in evolutionary biology. It demands accurate phylogenetic inference for both extant and extinct organisms, the latter being almost entirely dependent on morphological data. While parsimony methods have traditionally dominated the field of morphological phylogenetics, a rapidly growing number of studies are now employing probabilistic methods (maximum likelihood and Bayesian inference). The present-day toolkit of probabilistic methods offers varied software with distinct algorithms and assumptions for reaching global optimality. However, benchmark performance assessments of different software packages for the analyses of morphological data, particularly in the era of big data, are still lacking. Here, we test the performance of four major probabilistic software under variable taxonomic sampling and missing data conditions: the Bayesian inference-based programs MrBayes and RevBayes, and the maximum likelihood-based IQ-TREE and RAxML. We evaluated software performance by calculating the distance between inferred and true trees using a variety of metrics, including Robinson-Foulds (RF), Matching Splits (MS), and Kuhner-Felsenstein (KF) distances. Our results show that increased taxonomic sampling improves accuracy, precision, and resolution of reconstructed topologies across all tested probabilistic software applications and all levels of missing data. Under the RF metric, Bayesian inference applications were the most consistent, accurate, and robust to variation in taxonomic sampling in all tested conditions, especially at high levels of missing data, with little difference in performance between the two tested programs. The MS metric favored more resolved topologies that were generally produced by IQ-TREE. Adding more taxa dramatically reduced performance disparities between programs. Importantly, our results suggest that the RF metric penalizes incorrectly resolved nodes (false positives) more severely than the MS metric, which instead tends to penalize polytomies. If false positives are to be avoided in systematics, Bayesian inference should be preferred over maximum likelihood for the analysis of morphological data.

Morphological Phylogenetics Evaluated Using Novel Evolutionary Simulations

Evolutionary inferences require reliable phylogenies. Morphological data have traditionally been analyzed using maximum parsimony, but recent simulation studies have suggested that Bayesian analyses yield more accurate trees. This debate is ongoing, in part, because of ambiguity over modes of morphological evolution and a lack of appropriate models. Here, we investigate phylogenetic methods using two novel simulation models-one in which morphological characters evolve stochastically along lineages and another in which individuals undergo selection. Both models generate character data and lineage splitting simultaneously: the resulting trees are an emergent property, rather than a fixed parameter. Standard consensus methods for Bayesian searches (Mki) yield fewer incorrect nodes and quartets than the standard consensus trees recovered using equal weighting and implied weighting parsimony searches. Distances between the pool of derived trees (most parsimonious or posterior distribution) and the true trees-measured using Robinson-Foulds (RF), subtree prune and regraft (SPR), and tree bisection reconnection (TBR) metrics-demonstrate that this is related to the search strategy and consensus method of each technique. The amount and structure of homoplasy in character data differ between models. Morphological coherence, which has previously not been considered in this context, proves to be a more important factor for phylogenetic accuracy than homoplasy. Selection-based models exhibit relatively lower homoplasy, lower morphological coherence, and higher inaccuracy in inferred trees. Selection is a dominant driver of morphological evolution, but we demonstrate that it has a confounding effect on numerous character properties which are fundamental to phylogenetic inference. We suggest that the current debate should move beyond considerations of parsimony versus Bayesian, toward identifying modes of morphological evolution and using these to build models for probabilistic search methods. [Bayesian; evolution; morphology; parsimony; phylogenetics; selection; simulation.].

Bayesian Tip-Dated Phylogenetics in Paleontology: Topological Effects and Stratigraphic Fit

The incorporation of stratigraphic data into phylogenetic analysis has a long history of debate but is not currently standard practice for paleontologists. Bayesian tip-dated (or morphological clock) phylogenetic methods have returned these arguments to the spotlight, but how tip dating affects the recovery of evolutionary relationships has yet to be fully explored. Here I show, through analysis of several data sets with multiple phylogenetic methods, that topologies produced by tip dating are outliers as compared to topologies produced by parsimony and undated Bayesian methods, which retrieve broadly similar trees. Unsurprisingly, trees recovered by tip dating have better fit to stratigraphy than trees recovered by other methods under both the Gap Excess Ratio (GER) and the Stratigraphic Completeness Index (SCI). This is because trees with better stratigraphic fit are assigned a higher likelihood by the fossilized birth-death tree model. However, the degree to which the tree model favors tree topologies with high stratigraphic fit metrics is modulated by the diversification dynamics of the group under investigation. In particular, when net diversification rate is low, the tree model favors trees with a higher GER compared to when net diversification rate is high. Differences in stratigraphic fit and tree topology between tip dating and other methods are concentrated in parts of the tree with weaker character signal, as shown by successive deletion of the most incomplete taxa from two data sets. These results show that tip dating incorporates stratigraphic data in an intuitive way, with good stratigraphic fit an expectation that can be overturned by strong evidence from character data. [fossilized birth-death; fossils; missing data; morphological clock; morphology; parsimony; phylogenetics.].

Which morphological characters are influential in a Bayesian phylogenetic analysis? Examples from the earliest osteichthyans

There has been much recent debate about which method is best for reconstructing the tree of life from morphological datasets. However, little attention has been paid to which characters, if any, are responsible for topological differences between trees recovered from competing methods on empirical datasets. Indeed, a simple procedure for finding characters supporting conflicting tree topologies is available in a parsimony framework, but an equivalent procedure in a model-based framework is lacking. Here, I introduce such a procedure and apply it to the problem of the 'psarolepid' osteichthyans. The 'psarolepids', which include the earliest known osteichthyans, are weakly supported as stem osteichthyans under parsimony but strongly supported as sarcopterygians in Bayesian analysis. The Bayesian result is driven by just two characters, both of which relate to the intracranial joint of sarcopterygians. Important characters that support a stem osteichthyan affinity for 'psarolepids', such as the absence of tooth enamel, have virtually no effect in a Bayesian framework. This is because of a bias towards characters with relatively complete sampling, a bias that has previously been reported for molecular data. This has important implications for Bayesian analysis of morphological datasets in general, as characters from different body parts commonly have different levels of coding completeness. Methods to critically appraise character support for conflicting phylogenetic hypotheses, such as that used here, should form an important part of phylogenetic analyses.

Bayesian and parsimony approaches reconstruct informative trees from simulated morphological datasets

Phylogenetic analysis aims to establish the true relationships between taxa. Different analytical methods, however, can reach different conclusions. In order to establish which approach best reconstructs true relationships, previous studies have simulated datasets from known tree topologies, and identified the method that reconstructs the generative tree most accurately. On this basis, researchers have argued that morphological datasets should be analysed by Bayesian approaches, which employ an explicit probabilistic model of evolution, rather than parsimony methods-with implied weights parsimony sometimes identified as particularly inaccurate. Accuracy alone, however, is an inadequate measure of a tree's utility: a fully unresolved tree is perfectly accurate, yet contains no phylogenetic information. The highly resolved trees recovered by implied weights parsimony in fact contain as much useful information as the more accurate, but less resolved, trees recovered by Bayesian methods. By collapsing poorly supported groups, this superior resolution can be traded for accuracy, resulting in trees as accurate as those obtained by a Bayesian approach. By contrast, equally weighted parsimony analysis produces trees that are less resolved and less accurate, leading to less reliable evolutionary conclusions.

Data partitioning and correction for ascertainment bias reduce the uncertainty of placental mammal divergence times inferred from the morphological clock

Bayesian estimates of divergence times based on the molecular clock yield uncertainty of parameter estimates measured by the width of posterior distributions of node ages. For the relaxed molecular clock, previous works have reported that some of the uncertainty inherent to the variation of rates among lineages may be reduced by partitioning data. Here we test this effect for the purely morphological clock, using placental mammals as a case study. We applied the uncorrelated lognormal relaxed clock to morphological data of 40 extant mammalian taxa and 4,533 characters, taken from the largest published matrix of discrete phenotypic characters. The morphologically derived timescale was compared to divergence times inferred from molecular and combined data. We show that partitioning data into anatomical units significantly reduced the uncertainty of divergence time estimates for morphological data. For the first time, we demonstrate that ascertainment bias has an impact on the precision of morphological clock estimates. While analyses including molecular data suggested most divergences between placental orders occurred near the K-Pg boundary, the partitioned morphological clock recovered older interordinal splits and some younger intraordinal ones, including significantly later dates for the radiation of bats and rodents, which accord to the short-fuse hypothesis.

Phylogeny of Paleozoic limbed vertebrates reassessed through revision and expansion of the largest published relevant data matrix

The largest published phylogenetic analysis of early limbed vertebrates (Ruta M, Coates MI. 2007. Journal of Systematic Palaeontology 5:69-122) recovered, for example, Seymouriamorpha, Diadectomorpha and (in some trees) Caudata as paraphyletic and found the "temnospondyl hypothesis" on the origin of Lissamphibia (TH) to be more parsimonious than the "lepospondyl hypothesis" (LH)-though only, as we show, by one step. We report 4,200 misscored cells, over half of them due to typographic and similar accidental errors. Further, some characters were duplicated; some had only one described state; for one, most taxa were scored after presumed relatives. Even potentially continuous characters were unordered, the effects of ontogeny were not sufficiently taken into account, and data published after 2001 were mostly excluded. After these issues are improved-we document and justify all changes to the matrix-but no characters are added, we find (Analysis R1) much longer trees with, for example, monophyletic Caudata, Diadectomorpha and (in some trees) Seymouriamorpha; Ichthyostega either crownward or rootward of Acanthostega; and Anthracosauria either crownward or rootward of Temnospondyli. The LH is nine steps shorter than the TH (R2; constrained) and 12 steps shorter than the "polyphyly hypothesis" (PH-R3; constrained). Brachydectes (Lysorophia) is not found next to Lissamphibia; instead, a large clade that includes the adelogyrinids, urocordylid "nectrideans" and aïstopods occupies that position. As expected from the taxon/character ratio, most bootstrap values are low. Adding 56 terminal taxa to the original 102 increases the resolution (and decreases most bootstrap values). The added taxa range in completeness from complete articulated skeletons to an incomplete lower jaw. Even though the lissamphibian-like temnospondyls Gerobatrachus, Micropholis and Tungussogyrinus and the extremely peramorphic salamander Chelotriton are added, the difference between LH (R4; unconstrained) and TH (R5) rises to 10 steps, that between LH and PH (R6) to 15; the TH also requires several more regains of lost bones than the LH. Casineria, in which we tentatively identify a postbranchial lamina, emerges rather far from amniote origins in a gephyrostegid-chroniosuchian grade. Bayesian inference (Analysis EB, settings as in R4) mostly agrees with R4. High posterior probabilities are found for Lissamphibia (1.00) and the LH (0.92); however, many branches remain weakly supported, and most are short, as expected from the small character sample. We discuss phylogeny, approaches to coding, methods of phylogenetics (Bayesian inference vs. equally weighted vs. reweighted parsimony), some character complexes (e.g. preaxial/postaxial polarity in limb development), and prospects for further improvement of this matrix. Even in its revised state, the matrix cannot provide a robust assessment of the phylogeny of early limbed vertebrates. Sufficient improvement will be laborious-but not difficult.

Probabilistic methods surpass parsimony when assessing clade support in phylogenetic analyses of discrete morphological data

Fossil taxa are critical to inferences of historical diversity and the origins of modern biodiversity, but realizing their evolutionary significance is contingent on restoring fossil species to their correct position within the tree of life. For most fossil species, morphology is the only source of data for phylogenetic inference; this has traditionally been analysed using parsimony, the predominance of which is currently challenged by the development of probabilistic models that achieve greater phylogenetic accuracy. Here, based on simulated and empirical datasets, we explore the relative efficacy of competing phylogenetic methods in terms of clade support. We characterize clade support using bootstrapping for parsimony and Maximum Likelihood, and intrinsic Bayesian posterior probabilities, collapsing branches that exhibit less than 50% support. Ignoring node support, Bayesian inference is the most accurate method in estimating the tree used to simulate the data. After assessing clade support, Bayesian and Maximum Likelihood exhibit comparable levels of accuracy, and parsimony remains the least accurate method. However, Maximum Likelihood is less precise than Bayesian phylogeny estimation, and Bayesian inference recaptures more correct nodes with higher support compared to all other methods, including Maximum Likelihood. We assess the effects of these findings on empirical phylogenies. Our results indicate probabilistic methods should be favoured over parsimony.

Probabilistic methods surpass parsimony when assessing clade support in phylogenetic analyses of discrete morphological data

Fossil taxa are critical to inferences of historical diversity and the origins of modern biodiversity, but realizing their evolutionary significance is contingent on restoring fossil species to their correct position within the tree of life. For most fossil species, morphology is the only source of data for phylogenetic inference; this has traditionally been analysed using parsimony, the predominance of which is currently challenged by the development of probabilistic models that achieve greater phylogenetic accuracy. Here, based on simulated and empirical datasets, we explore the relative efficacy of competing phylogenetic methods in terms of clade support. We characterize clade support using bootstrapping for parsimony and Maximum Likelihood, and intrinsic Bayesian posterior probabilities, collapsing branches that exhibit less than 50% support. Ignoring node support, Bayesian inference is the most accurate method in estimating the tree used to simulate the data. After assessing clade support, Bayesian and Maximum Likelihood exhibit comparable levels of accuracy, and parsimony remains the least accurate method. However, Maximum Likelihood is less precise than Bayesian phylogeny estimation, and Bayesian inference recaptures more correct nodes with higher support compared to all other methods, including Maximum Likelihood. We assess the effects of these findings on empirical phylogenies. Our results indicate probabilistic methods should be favoured over parsimony.