CONDUCTING PHYLOGENETIC COMPARATIVE STUDIES WHEN THE PHYLOGENY IS NOT KNOWN

A method is proposed to conduct phylogenetic analyses of comparative or interspecific data when the true phylogeny is not known. Standard models of speciation and/or extinction or other methods are used to generate a sample from the set of all possible phylogenies for the measured species. The comparative data are then analyzed on each of the possible trees to obtain a distribution of possible evolutionary statistics for these data. The mean of this distribution is proposed as a reasonable estimate of the true evolutionary statistic of interest. Ways of obtaining confidence intervals and of developing hypothesis tests for this mean statistic are also proposed. The method can be used with any comparative method or phylogenetic analysis technique when phylogenetic relationships among species are not known or when branch lengths for a phylogeny in units of expected character change (as required by most methods) are not available. Computer programs to conduct the analyses are available on request.     

Assessing among‐lineage variability in phylogenetic imputation of functional trait datasets

Phylogenetic imputation has recently emerged as a potentially powerful tool for predicting missing data in functional traits datasets. As such, understanding the limitations of phylogenetic modelling in predicting trait values is critical if we are to use them in subsequent analyses. Previous studies have focused on the relationship between phylogenetic signal and clade‐level prediction accuracy, yet variability in prediction accuracy among individual tips of phylogenies remains largely unexplored. Here, we used simulations of trait evolution along the branches of phylogenetic trees to show how the accuracy of phylogenetic imputations is influenced by the combined effects of 1) the amount of phylogenetic signal in the traits and 2) the branch length of the tips to be imputed. Specifically, we conducted cross‐validation trials to estimate the variability in prediction accuracy among individual tips on the phylogenies (hereafter ‘tip‐level accuracy’). We found that under a Brownian motion model of evolution (BM, Pagel't λ = 1), tip‐level accuracy rapidly decreased with increasing tip branch‐lengths, and only tips of approximately 10% or less of the total height of the trees showed consistently accurate predictions (i.e. cross‐validation R‐squared >0.75). When phylogenetic signal was weak, the effect of tip branch‐length was reduced, becoming negligible for traits simulated with λ < 0.7, where accuracy was in any case low. Our study shows that variability in prediction accuracy among individual tips of the phylogeny should be considered when evaluating the reliability of phylogenetically imputed trait values. To address this challenge, we describe a Monte Carlo‐based method that allows one to estimate the expected tip‐level accuracy of phylogenetic predictions for continuous traits. Our approach identifies gaps in functional trait datasets for which phylogenetic imputation performs poorly, and will help ecologists to design more efficient trait collection campaigns by focusing resources on lineages whose trait values are more uncertain.     

Positive correlations between molecular and morphological rates of evolution

The existence of positive associations between rates of molecular and morphological evolution (calculated from branch lengths of phylogenetic trees reconstructed using molecular and morphological characters, respectively) is important to issues of neutrality in sequence evolution, phylogenetic reconstructions assuming neutrality, and evolutionary genotype-phenotype mapping. Rates correlate positively when including branches leading to extant species (tips). Excluding tips, trends are similar, but statistical significances decrease systematically. This is due to (a) lower statistical power (excluding tips reduces sample sizes), and (b) rates are solely calculated from inaccurately reconstructed character states of extinct ancestral species, and this noise decreases correlation strengths. Correlations between molecular and morphological rates of evolution increase as more morphological characters are included for phylogenetic reconstruction. Sequence lengths apparently affect correlations along similar principles. Analyses of plant phylogenies confirm those from animals: sampling biases decrease correlations between molecular and morphological rates of evolution. Results confirm that genotype and phenotype are linked, and suggest adaptive components for molecular evolution. The discussion stresses the difficulties associated with analyses and conclusions based on data deduced from phylogenetic reconstruction.     

Directional biases in phylogenetic structure quantification: a Mediterranean case study

Recent years have seen an increasing effort to incorporate phylogenetic hypotheses to the study of community assembly processes. The incorporation of such evolutionary information has been eased by the emergence of specialized software for the automatic estimation of partially resolved supertrees based on published phylogenies. Despite this growing interest in the use of phylogenies in ecological research, very few studies have attempted to quantify the potential biases related to the use of partially resolved phylogenies and to branch length accuracy, and no work has examined how tree shape may affect inference of community phylogenetic metrics. In this study, we tested the influence of phylogenetic resolution and branch length information on the quantification of phylogenetic structure, and also explored the impact of tree shape (stemminess) on the loss of accuracy in phylogenetic structure quantification due to phylogenetic resolution. For this purpose, we used 9 sets of phylogenetic hypotheses of varying resolution and branch lengths to calculate three indices of phylogenetic structure: the mean phylogenetic distance (NRI), the mean nearest taxon distance (NTI) and phylogenetic diversity (stdPD) metrics. The NRI metric was the less sensitive to phylogenetic resolution, stdPD showed an intermediate sensitivity, and NTI was the most sensitive one; NRI was also less sensitive to branch length accuracy than NTI and stdPD, the degree of sensitivity being strongly dependent on the dating method and the sample size. Directional biases were generally towards type II errors. Interestingly, we detected that tree shape influenced the accuracy loss derived from the lack of phylogenetic resolution, particularly for NRI and stdPD. We conclude that well‐resolved molecular phylogenies with accurate branch length information are needed to identify the underlying phylogenetic structure of communities, and also that sensitivity of phylogenetic structure measures to low phylogenetic resolution can strongly vary depending on phylogenetic tree shape.     

FiSSE: A simple nonparametric test for the effects of a binary character on lineage diversification rates

It is widely assumed that phenotypic traits can influence rates of speciation and extinction, and several statistical approaches have been used to test for correlations between character states and lineage diversification. Recent work suggests that model‐based tests of state‐dependent speciation and extinction are sensitive to model inadequacy and phylogenetic pseudoreplication. We describe a simple nonparametric statistical test (“FiSSE”) to assess the effects of a binary character on lineage diversification rates. The method involves computing a test statistic that compares the distributions of branch lengths for lineages with and without a character state of interest. The value of the test statistic is compared to a null distribution generated by simulating character histories on the observed phylogeny. Our tests show that FiSSE can reliably infer trait‐dependent speciation on phylogenies of several hundred tips. The method has low power to detect trait‐dependent extinction but can infer state‐dependent differences in speciation even when net diversification rates are constant. We assemble a range of macroevolutionary scenarios that are problematic for likelihood‐based methods, and we find that FiSSE does not show similarly elevated false positive rates. We suggest that nonparametric statistical approaches, such as FiSSE, provide an important complement to formal process‐based models for trait‐dependent diversification.     

On the selection of phylogenetic eigenvectors for ecological analyses

Among the statistical methods available to control for phylogenetic autocorrelation in ecological data, those based on eigenfunction analysis of the phylogenetic distance matrix among the species are becoming increasingly important tools. Here, we evaluate a range of criteria to select eigenvectors extracted from a phylogenetic distance matrix (using phylogenetic eigenvector regression, PVR) that can be used to measure the level of phylogenetic signal in ecological data and to study correlated evolution. We used a principal coordinate analysis to represent the phylogenetic relationships among 209 species of Carnivora by a series of eigenvectors, which were then used to model log‐transformed body size. We first conducted a series of PVRs in which we increased the number of eigenvectors from 1 to 70, following the sequence of their associated eigenvalues. Second, we also investigated three non‐sequential approaches based on the selection of 1) eigenvectors significantly correlated with body size, 2) eigenvectors selected by a standard stepwise algorithm, and 3) the combination of eigenvectors that minimizes the residual phylogenetic autocorrelation. We mapped the mean specific component of body size to evaluate how these selection criteria affect the interpretation of non‐phylogenetic signal in Bergmann's rule. For comparison, the same patterns were analyzed using autoregressive model (ARM) and phylogenetic generalized least‐squares (PGLS). Despite the robustness of PVR to the specific approaches used to select eigenvectors, using a relatively small number of eigenvectors may be insufficient to control phylogenetic autocorrelation, leading to flawed conclusions about patterns and processes. The method that minimizes residual autocorrelation seems to be the best choice according to different criteria. Thus, our analyses show that, when the best criterion is used to control phylogenetic structure, PVR can be a valuable tool for testing hypotheses related to heritability at the species level, phylogenetic niche conservatism and correlated evolution between ecological traits.     

The evolution of body size, Cope's rule and the origin of amniotes

The evolution of body size in tetrapods is assessed using a database that includes 107 early stegocephalian species ranging in time from the Frasnian (Upper Devonian) to the Tatarian (Upper Permian). All analyses use methods that incorporate phylogenetic information (topology and branch lengths). In all tests, the impact of alternative topologies and branch lengths are assessed. Previous reports that raised doubts about the accuracy of squared-change parsimony assessment of ancestral character value appear to have used datasets in which there was no phylogenetic signal. Hence, squared-change parsimony may be more reliable than suggested in recent studies, at least when a phylogenetic signal is present in the datasets of interest. Analysis using random taxon reshuffling on three reference phylogenies shows that cranial and presacral length include a strong phylogenetic signal. Character optimization of body size in stegocephalians using squared-change parsimony on a time-calibrated phylogeny incorporating branch length information is used to test a previously published scenario on the origin of amniotes and of the amniotic egg that implies that the ancestors of amniotes were small (no more than 10 cm in snout-vent length), and that their size increased subsequent to the appearance of the amniotic egg. The optimization suggests that first amniotes were somewhat larger than previously hypothesized; the estimated snout-vent length is about 24 cm, and the lower end of the 95% confidence interval of the phylogeny that yields the smallest inferred size suggests that no ancestor of amniotes measured less than 12 cm in snout-vent length. Character optimization, permutational multiple linear regressions, and independent contrast analyses show that Cope's rule of phyletic size increase applies to early reptiliomorphs but that it does not apply to early stegocephalians globally.     

Bayesian estimation of ancestral character states on phylogenies

Biologists frequently attempt to infer the character states at ancestral nodes of a phylogeny from the distribution of traits observed in contemporary organisms. Because phylogenies are normally inferences from data, it is desirable to account for the uncertainty in estimates of the tree and its branch lengths when making inferences about ancestral states or other comparative parameters. Here we present a general Bayesian approach for testing comparative hypotheses across statistically justified samples of phylogenies, focusing on the specific issue of reconstructing ancestral states. The method uses Markov chain Monte Carlo techniques for sampling phylogenetic trees and for investigating the parameters of a statistical model of trait evolution. We describe how to combine information about the uncertainty of the phylogeny with uncertainty in the estimate of the ancestral state. Our approach does not constrain the sample of trees only to those that contain the ancestral node or nodes of interest, and we show how to reconstruct ancestral states of uncertain nodes using a most-recent-common-ancestor approach. We illustrate the methods with data on ribonuclease evolution in the Artiodactyla. Software implementing the methods (BayesMultiState) is available from the authors.     

DISTANCE METHODS: A REPLY TO FARRIS

Abstract— Farris (1985) claimed that my assertions about unbiasedness and consistency of estimates of a phylogeny obtained by least squares fitting are in error. The counterexample he constructed violates the assumptions of additivity and independence of distances which were clearly stated in my earlier paper. As such it is not a valid counterexample. It is argued, contrary to Farris's claims, that one need not avoid nonmetric distances, and that one should avoid negative branch lengths in estimates of phylogenies from distance data. Statistical tests of clockness, and, to a limited extent, of alternative phylogenies can be constructed, and these are demonstrated by example. A computer program to infer phylogenies from distance matrices has been in free distribution by me for several years; it seems as effective as the program recently announced by Farris. Information on phylogenies is present in distance data, as in other kinds of data, and statistical methods can be developed to extract it.     

TESTING FOR PHYLOGENETIC SIGNAL IN BIOLOGICAL TRAITS: THE UBIQUITY OF CROSS‐PRODUCT STATISTICS

To evaluate rates of evolution, to establish tests of correlation between two traits, or to investigate to what degree the phylogeny of a species assemblage is predictive of a trait value so‐called tests for phylogenetic signal are used. Being based on different approaches, these tests are generally thought to possess quite different statistical performances. In this article, we show that the Blomberg et al. K and K*, the Abouheif index, the Moran's I, and the Mantel correlation are all based on a cross‐product statistic, and are thus all related to each other when they are associated to a permutation test of phylogenetic signal. What changes is only the way phylogenetic and trait similarities (or dissimilarities) among the tips of a phylogeny are computed. The definitions of the phylogenetic and trait‐based (dis)similarities among tips thus determines the performance of the tests. We shortly discuss the biological and statistical consequences (in terms of power and type I error of the tests) of the observed relatedness among the statistics that allow tests for phylogenetic signal. Blomberg et al. K* statistic appears as one on the most efficient approaches to test for phylogenetic signal. When branch lengths are not available or not accurate, Abouheif's C_mean statistic is a powerful alternative to K*.     

Eumalacostracan phylogeny and total evidence: limitations of the usual suspects

Background  

The phylogeny of Eumalacostraca (Crustacea) remains elusive, despite over a century of interest. Recent morphological and molecular phylogenies appear highly incongruent, but this has not been assessed quantitatively. Moreover, 18S rRNA trees show striking branch length differences between species, accompanied by a conspicuous clustering of taxa with similar branch lengths. Surprisingly, previous research found no rate heterogeneity. Hitherto, no phylogenetic analysis of all major eumalacostracan taxa (orders) has either combined evidence from multiple loci, or combined molecular and morphological evidence.     

Modeling body size evolution in Felidae under alternative phylogenetic hypotheses

The use of phylogenetic comparative methods in ecological research has advanced during the last twenty years, mainly due to accurate phylogenetic reconstructions based on molecular data and computational and statistical advances. We used phylogenetic correlograms and phylogenetic eigenvector regression (PVR) to model body size evolution in 35 worldwide Felidae (Mammalia, Carnivora) species using two alternative phylogenies and published body size data. The purpose was not to contrast the phylogenetic hypotheses but to evaluate how analyses of body size evolution patterns can be affected by the phylogeny used for comparative analyses (CA). Both phylogenies produced a strong phylogenetic pattern, with closely related species having similar body sizes and the similarity decreasing with increasing distances in time. The PVR explained 65% to 67% of body size variation and all Moran's I values for the PVR residuals were non-significant, indicating that both these models explained phylogenetic structures in trait variation. Even though our results did not suggest that any phylogeny can be used for CA with the same power, or that "good" phylogenies are unnecessary for the correct interpretation of the evolutionary dynamics of ecological, biogeographical, physiological or behavioral patterns, it does suggest that developments in CA can, and indeed should, proceed without waiting for perfect and fully resolved phylogenies.     

Exploring patterns of interspecific variation in quantitative traits using sequential phylogenetic eigenvector regressions

A number of metrics have been developed for estimating phylogenetic signal in data and to evaluate correlated evolution, inferring broad-scale evolutionary and ecological processes. Here, we proposed an approach called phylogenetic signal-representation (PSR) curve, built upon phylogenetic eigenvector regression (PVR). In PVR, selected eigenvectors extracted from a phylogenetic distance matrix are used to model interspecific variation. In the PSR curve, sequential PVR models are fitted after successively increasing the number of eigenvectors and plotting their R(2) against the accumulated eigenvalues. We used simulations to show that a linear PSR curve is expected under Brownian motion and that its shape changes under alternative evolutionary models. The PSR area, expressing deviations from Brownian motion, is strongly correlated (r= 0.873; P < 0.01) with Blomberg's K-statistics, so nonlinear PSR curves reveal if traits are evolving at a slower or higher rate than expected by Brownian motion. The PSR area is also correlated with phylogenetic half-life under an Ornstein-Uhlenbeck process, suggesting how both methods describe the shape of the relationship between interspecific variation and time since divergence among species. The PSR curve provides an elegant exploratory method to understand deviations from Brownian motion, in terms of acceleration or deceleration of evolutionary rates occurring at large or small phylogenetic distances.     

Detecting the node-density artifact in phylogeny reconstruction

The node-density effect is an artifact of phylogeny reconstruction that can cause branch lengths to be underestimated in areas of the tree with fewer taxa. Webster, Payne, and Pagel (2003, Science 301:478) introduced a statistical procedure (the "delta" test) to detect this artifact, and here we report the results of computer simulations that examine the test's performance. In a sample of 50,000 random data sets, we find that the delta test detects the artifact in 94.4% of cases in which it is present. When the artifact is not present (n = 10,000 simulated data sets) the test showed a type I error rate of approximately 1.69%, incorrectly reporting the artifact in 169 data sets. Three measures of tree shape or "balance" failed to predict the size of the node-density effect. This may reflect the relative homogeneity of our randomly generated topologies, but emphasizes that nearly any topology can suffer from the artifact, the effect not being confined only to highly unevenly sampled or otherwise imbalanced trees. The ability to screen phylogenies for the node-density artifact is important for phylogenetic inference and for researchers using phylogenetic trees to infer evolutionary processes, including their use in molecular clock dating.     

Measuring Asymmetry in Time-Stamped Phylogenies

Previous work has shown that asymmetry in viral phylogenies may be indicative of heterogeneity in transmission, for example due to acute HIV infection or the presence of ‘core groups’ with higher contact rates. Hence, evidence of asymmetry may provide clues to underlying population structure, even when direct information on, for example, stage of infection or contact rates, are missing. However, current tests of phylogenetic asymmetry (a) suffer from false positives when the tips of the phylogeny are sampled at different times and (b) only test for global asymmetry, and hence suffer from false negatives when asymmetry is localised to part of a phylogeny. We present a simple permutation-based approach for testing for asymmetry in a phylogeny, where we compare the observed phylogeny with random phylogenies with the same sampling and coalescence times, to reduce the false positive rate. We also demonstrate how profiles of measures of asymmetry calculated over a range of evolutionary times in the phylogeny can be used to identify local asymmetry. In combination with different metrics of asymmetry, this combined approach offers detailed insights of how phylogenies reconstructed from real viral datasets may deviate from the simplistic assumptions of commonly used coalescent and birth-death process models.     

Building megaphylogenies for macroecology: taking up the challenge

The last decades have seen an upsurge in ecological studies incorporating phylogenetic information with increasing species samples, motivated by the common conjecture that species with common ancestors should share some ecological characteristics due to niche conservatism. This has been carried out using various methods of increasing complexity and reliability: using only taxonomical classification; constructing supertrees that incorporate only topological information from previously published phylogenies; or building supermatrices of molecular data that are used to estimate phylogenies with evolutionary meaningful branch lengths. Although the latter option is more informative than the others, it remains under‐used in ecology because ecologists are generally unaware of or unfamiliar with modern molecular phylogenetic methods. However, a solid phylogenetic hypothesis is necessary to conduct reliable ecological analysis integrating evolutive aspects. Our aim here is to clarify the concepts and methodological issues associated with the reconstruction of dated megaphylogenies, and to show that it is nowadays possible to obtain accurate and well sampled megaphylogenies with informative branch‐lengths on large species samples. This is possible thanks to improved phylogenetic methods, vast amounts of molecular data available from databases such as Genbank, and consensus knowledge on deep phylogenetic relationships for an increasing number of groups of organisms. Finally, we include a detailed step‐by‐step workflow pipeline (Supplementary material), from data acquisition to phylogenetic inference, mainly based on the R environment (widely used by ecologists) and the use of free web‐servers, that has been applied to the reconstruction of a species‐level phylogeny of all breeding birds of Europe.     

A novel algorithm and web-based tool for comparing two alternative phylogenetic trees

SUMMARY: We describe an algorithm and software tool for comparing alternative phylogenetic trees. The main application of the software is to compare phylogenies obtained using different phylogenetic methods for some fixed set of species or obtained using different gene sequences from those species. The algorithm pairs up each branch in one phylogeny with a matching branch in the second phylogeny and finds the optimum 1-to-1 map between branches in the two trees in terms of a topological score. The software enables the user to explore the corresponding mapping between the phylogenies interactively, and clearly highlights those parts of the trees that differ, both in terms of topology and branch length. AVAILABILITY: The software is implemented as a Java applet at http://www.mrc-bsu.cam.ac.uk/personal/thomas/phylo_comparison/comparison_page.html. It is also available on request from the authors.     

BEST-FIT MAXIMUM-LIKELIHOOD MODELS FOR PHYLOGENETIC INFERENCE: EMPIRICAL TESTS WITH KNOWN PHYLOGENIES

Despite the proliferation of increasingly sophisticated models of DNA sequence evolution, choosing among models remains a major problem in phylogenetic reconstruction. The choice of appropriate models is thought to be especially important when there is large variation among branch lengths. We evaluated the ability of nested models to reconstruct experimentally generated, known phylogenies of bacteriophage T7 as we varied the terminal branch lengths. Then, for each phylogeny we determined the best-fit model by progressively adding parameters to simpler models. We found that in several cases the choice of best-fit model was affected by the parameter addition sequence. In terms of phylogenetic performance, there was little difference between models when the ratio of short: long terminal branches was 1:3 or less. However, under conditions of extreme terminal branch-length variation, there were not only dramatic differences among models, but best-fit models were always among the best at overcoming long-branch attraction. The performance of minimum-evolution-distance methods was generally lower than that of discrete maximum-likelihood methods, even if maximum-likelihood methods were used to generate distance matrices. Correcting for among-site rate variation was especially important for overcoming long-branch attraction. The generality of our conclusions is supported by earlier simulation studies and by a preliminary analysis of mitochondrial and nuclear sequences from a well-supported four-taxon amniote phylogeny.     

Is cladogenesis heritable?

The heritability of speciation rates and extinction risks is a crucial parameter in models of macroevolution, but little direct evidence has been available to assess the occurrence, strength, or generality of this heritability. We tested for heritability using correlations between ancestral and descendent branch lengths in phylogenetic trees, an approach first applied to a bird phylogeny by Harvey et al. (1991, pages 123-137 in Genes in ecology [R. J. Berry et al., eds.], Blackwell Scientific, Oxford). We applied Harvey et al.'s test to some of the largest DNA sequence-based phylogenetic analyses published to date for plants, insects, fungi, and bacteria. If one of two parent lineages splits first and if this is the case for any heritable reason, then on average we expect its daughter lineages to also split first. We also used a randomization procedure to assess significance of branch length heritability. Using maximum parsimony and maximum likelihood branch lengths and trees made ultrametric after nonparametric rate smoothing or by enforcing a molecular clock, we found a pattern for most clades consistent with heritable net cladogenesis. Heritability of cladogenesis may be a general phenomenon, detectable across a large number of lineages and a broad range of taxa.