AN EMPIRICAL COMPARISON OF MICROCOMPUTER PARSIMONY PROGRAMS

Abstract— The effectiveness and efficiency of two microcomputer parsimony programs (Felsenstein's PHYLIP and Swofford's PAUP), and a prototype version of a third (SHEN, to be incorporated in Farris' HENNIG-86), are evaluated with reference to 35 data sets, including those previously used by Luckow and Pimentel (1985) to benchmark mainframe programs. Both PHYLIP and PAUP can be used effectively; with careful selection of options, their accuracy can equal or surpass that of older mainframe programs. PHYLIP is relatively inefficient (in use of computer time); its usefulness is also limited by the inability of its heuristic algorithms to detect multiple equally parsimonious solutions in a single run. PHYLIP's exact algorithm (branch-and-bound) and PAUP's heuristic and exact algorithms do not share the latter drawback but require users to spend unreasonable amounts of time coping with redundant output. The remarkable effectiveness of these programs offers hope that (particularly with the advent of multitasking microcomputers) exact and cost-effective solutions will be obtainable in many, if not all, cladistic analyses.     

IN SEARCH OF A METHOD FOR CLADISTIC BIOGEOGRAPHY: AN EMPIRICAL COMPARISON OF COMPONENT ANALYSIS, BROOKS PARSIMONY ANALYSIS, AND THREE-AREA STATEMENTS

Abstract— Three quantitative cladistic biogeographic methods, namely, component analysis, Brooks parsimony analysis (BPA), and three-area statements (TAS) have been proposed for obtaining general area cladograms from taxon-area cladograms. Available programs implementing these methods include COMPONENT versions 1.5 and 2.0 for component analysis, TAS for three-area statements, and Hennig86 for analysing matrices for both three-area statements and BPA. Ten different data sets were analysed with these programs and items of error were used to evaluate the general area cladograms obtained. None of the computer implementations of the methods compared proved to be more effective than the others.
"…all methodologies, even the most obvious ones, have their limits…"
Feyerabend (1993:231)     

ON MISSING ENTRIES IN CLADISTIC ANALYSIS

Abstract The exact algorithms of two commonly used parsimony programs, Hennig86 by J. S. Farris and PAUP by D. Swofford, sometimes produce different solutions, and sometimes produce resolutions that are not supported by the data being analysed. The discrepancies apparently involve the treatment of missing entries, which can currently represent unknown data, inapplicable character and/or polymorphic taxa. Each of those potential sources of ambiguity is logically (if not computationally) different; with regard to binary characters, unknown data could be either 0 or 1, inapplicable characters are neither 0 nor 1 and polymorphisms are both 0 and 1. Resolutions that cannot be supported by any possible combination of known state attributions should either be flagged as such or suppressed entirely.     

Effects of data incompleteness on the relative performance of parsimony and Bayesian approaches in a supermatrix phylogenetic reconstruction of Mustelidae and Procyonidae (Carnivora)

Missing data are commonly thought to impede a resolved or accurate reconstruction of phylogenetic relationships, and probabilistic analysis techniques are increasingly viewed as less vulnerable to the negative effects of data incompleteness than parsimony analyses. We test both assumptions empirically by conducting parsimony and Bayesian analyses on an approximately 1.5 × 10⁶‐cell (27 965 characters × 52 species) mustelid–procyonid molecular supermatrix with 62.7% missing entries. Contrary to the first assumption, phylogenetic relationships inferred from our analyses are fully (Bayesian) or almost fully (parsimony) resolved topologically with mostly strong support and also largely in accord with prior molecular estimations of mustelid and procyonid phylogeny derived with parsimony, Bayesian, and other probabilistic analysis techniques from smaller but complete or nearly complete data sets. Contrary to the second assumption, we found no compelling evidence in support of a relationship between the inferior performance of parsimony and taxon incompleteness (i.e. the proportion of missing character data for a taxon), although we found evidence for a connection between the inferior performance of parsimony and character incompleteness (i.e. no overlap in character data between some taxa). The relatively good performance of our analyses may be related to the large number of sampled characters, so that most taxa (even highly incomplete ones) are represented by a sufficient number of characters allowing both approaches to resolve their relationships. © The Willi Hennig Society 2009.     

The phylogeny of elateroidea (Coleoptera), or which tree is best today?

Abstract — The relationships of the clicking Elateroidea beetles were studied with the help of parsimony analysis using Hennig86. The character matrix included 70 characters and 27 taxa. The results demonstrate the monophyly of the group Throscidae sensu Crowson, contrary to views presented in other papers. Methods for solving this problem were sought. When several minimum length solutions were obtained, successive weighting and a search for a strict consensus tree identical with one of the original trees appeared to be acceptable ways for trying to identify the preferred solution. When conflicting trees from separate data sets were compared, a combined global analysis turned out to be impossible to perform because the data sets used different terminal taxa. In this case, the incongruence and total support tests provided by Farris' programs RNA and KON proved indispensable. The conflict found between the results obtained here and those presented by other workers using a large suite of larval characters were shown to be caused by an incongruent data matrix used in the latter study—the larval data set resulted in a polyphyletic ingroup and suggests relationships quite different from adult data alone. Directed large scale homoplasy due to repeated re-invasion of two major habitats by separate clades may be the factor causing difficulties in coding the larval characters.     

Systematics, cladistics and biogeography of the American genus Farrodes (Ephemeroptera: Leptophlebiidae: Atalophlebiinae)

Cladistic and biogeographic analyses of the genus Farrodes are presented. Two species groups are delineated within Farrodes: F. caribbianus and F. bimaculatus. Three species formerly placed in other genera– Thraulus caribbianus Traver, Thraulus roundsi Traver and Homothraulus maculatus (Needham & Murphy)–are transferred to Farrodes. The species of the F. caribbianus species group are revised. Three new species are described: F. savagei from Venezuela, F. maya and F. mexicanus from Mexico. Keys to separate the two species groups of Farrodes and the species of the F caribbianus species group are provided. Successive cladistic analyses were carried out on both adult and nymphal characters using Hennig86 and CLADOS. The matrix was composed of all available data (nymphal characters were missing for some species), from nymphal and adult stages separately and on taxa represented by both adult and nymphal characters. Species of the genera Simothraulopsis and Homothraulus (components of the Farrodes lineage) were included in the analyses, and Ecuaphkbia was used as the outgroup. Results of the four analyses are compatible. The historical biogeography of Farrodes , with a distribution from nordiern Argentina to southern Texas, is analysed using the program COMPONENT. Areas of endemism are established, and some of their relationships compared with those of other groups available in the literature.     

The relative performance of Bayesian and parsimony approaches when sampling characters evolving under homogeneous and heterogeneous sets of parameters

We tested whether it is beneficial for the accuracy of phylogenetic inference to sample characters that are evolving under different sets of parameters, using both Bayesian MCMC (Markov chain Monte Carlo) and parsimony approaches. We examined differential rates of evolution among characters, differential character-state frequencies and character-state space, and differential relative branch lengths among characters. We also compared the relative performance of parsimony and Bayesian analyses by progressively incorporating more of these heterogeneous parameters and progressively increasing the severity of this heterogeneity. Bayesian analyses performed better than parsimony when heterogeneous simulation parameters were incorporated into the substitution model. However, parsimony outperformed Bayesian MCMC when heterogeneous simulation parameters were not incorporated into the Bayesian substitution model. The higher the rate of evolution simulated, the better parsimony performed relative to Bayesian analyses. Bayesian and parsimony analyses converged in their performance as the number of simulated heterogeneous model parameters increased. Up to a point, rate heterogeneity among sites was generally advantageous for phylogenetic inference using both approaches. In contrast, branch-length heterogeneity was generally disadvantageous for phylogenetic inference using both parsimony and Bayesian approaches. Parsimony was found to be more conservative than Bayesian analyses, in that it resolved fewer incorrect clades.
© The Willi Hennig Society 2006.     

Phytogeny of the nematocerous families of Diptera (insecta)

The relationships of the nematocerous families of Diptera are cladistieally analysed using the parsimony programs PAUP and Hennig86. An extensive review, as well as a data matrix, is presented for 98 almost exclusively morphological characters (larva, 56; pupa, 6; adult, 36). Four infraorders are recognized, viz , Ptychopteromorpha, Culicomorpha, Blephariceromorpha, Bibionomorpha, and a clade containing the 'higher Nematocera' and Brachycera. Traditionally the family Nymphomyiidae or the infraorder Tipulomorpha (=Tipulidae, with or without Trichoceridae) are considered the most basal clade of the extant Diptera. On the basis of our cladistic analysis it is suggested that the Ptychopteromorpha-Culicomorpha clade is the sister-group of all other extant Diptera. We provide evidence that the Axymyiidae are part of a monophyletic Bibionomorpha. The latter infraorder is proposed as the sister-group of the higher Nematocera and Brachycera. We transfer the Tipulidae (Tipulomorpha) to the higher Nematocera, at a position next to Trichoceridae and near the Anisopodidae-Brachycera lineage. Previous hypotheses concerning nematocerous relationships are reviewed.     

Relationships within Podocarpaceae based on DNA sequence,anatomical, morphological,and biogeographical data

Despite considerable recent progress in understanding intergeneric relationships, a comprehensive analysis of Podocarpaceae at the species level using molecular data, biogeography, anatomy, and morphology has not been previously attempted. Here we present sequence analyses of rbcL, nrITS1 and NEEDLY intron 2 for two‐thirds (183 accessions of 145 taxa) of all Podocarpaceae species representing all genera except Parasitaxus. These analyses include many more species and accessions than previous studies and result in a more resolved phylogeny. The comprehensive anatomical and morphological study ensures that the identification of taxa is correct and also provides clade support. Bayesian and parsimony analyses were used to resolve 20 well‐supported monophyletic groups including 11 groups of the formerly poorly resolved subgenera Podocarpus and Foliolatus. The well‐resolved topology is supported by anatomical and morphological features and is highly congruent with geographical distribution. © The Willi Hennig Society 2011.     

Inference of phylogenetic relationships among key angiosperm lineages using a compatibility method on a molecular data set

Phylogenetic relationships among the five key angiosperm lineages,Ceratophyllum,Chloranthaceae,eudicots,magnoliids,and monocots,have resisted resolution despite several large-scale analyses sampling taxa and characters extensively and using various analytical methods.Meanwhile,compatibility methods,which were explored together with parsimony and likelihood methods during the early development stage of phylogenetics.have been greatly under-appreciated and not been used to analyze the massive amount of sequence data to reconstruct thye basal angiosperm phylogeny.In this study,we used a compatibility method on a data set of eight genes (mitochondrial atp1,matR,and nad5,plastid atpB,marK,rbcL,and rpoC2,and nuclear 18S rDNA)gathered in an earlier study.We selected two sets of characters that are compatible with more of the other characters than a random character would be with at probabilities of pM＜0.1 and p＜0.5 respectively.The resulting data matrices were subjected to parsimony and likelihood bootstrap analyses.Our unrooted parsimony analyses showed that Ceratophyllum was immediately related to eudicots,this larger lineage was immediately related to magnoliids,and monocots were closely related to Chloranthaceae.All these relationships received 76%-96% bootstrap support.A likelihood analysis of the 8 gene pM＜0.5 compatible site matrix recovered the same topology but with low support.Likelihood analyses of other compatible site matrices produced different topologies that were all weakly supported.The topology reconstructed in the parsimony analyses agrees with the one recovered in the previous study using both parsimony and likelihood methods when no character was eliminated.Parts of this topology have also been recovered in several earlier studies.Hence,this topology plausibly reflects the true relationships among the five key angiosperm lineages.     

Bias and conflict in phylogenetic inference of myco-heterotrophic plants: a case study in Thismiaceae

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. © The Willi Hennig Society 2009.     

Dynamic homology and the likelihood criterion

The use of likelihood as an optimality criterion is explored in the context of dynamic homology. Simple models and procedures are described to allow the analysis of large variable length sequence data sets, alone and in combination with qualitative information (such as morphology). Several approaches are discussed that have different likelihood interpretations in terms of maximum parsimony likelihood and maximum average likelihood. Implementation is discussed and an example in arthropod systematics presented. Topological congruence comparisons with parsimony are made.
© The Willi Hennig Society 2006.     

Unringing a bell: metazoan phylogenomics and the partition bootstrap

Several large phylogenomic analyses have recently cast doubt on long‐held beliefs about early metazoan phylogenetic patterns. Those data sets, and the relative bootstrap support for various controversial clades, are reanalysed in the context of parsimony, yielding results that are at considerable odds with the original likelihood or Bayesian findings. Discrepancies are considered in light of the tendency of RAxML to overestimate support values by virtue (sic) of its lazy search algorithm and its autocorrelated pseudoreplication as well as the extraordinary ability for Bayesian analyses to be led astray by missing data. In addition to standard nonparametric bootstrapping as a measure of support, a new strategy involving resampling loci as units, partition bootstrap support, is introduced as a more defensible alternative to resampling nonindependent sites. © The Willi Hennig Society 2009.     

Parsimony overcomes statistical inconsistency with the addition of more data from the same gene

Many authors have demonstrated that the parsimony method of phylogenetic analysis can fail to estimate phylogeny accurately under certain conditions when data follow a model that stipulates homogeneity of the evolutionary process. These demonstrations further show that no matter how much data are added, parsimony will forever exhibit this statistical inconsistency if the additional data have the same distributional properties as the original data. This final component—that the additional data must follow the same distribution as the original data—is crucial to the demonstration. Recent simulations show, however, that if data evolve heterogeneously, parsimony can perform consistently. Here we show, using natural data, that parsimony can overcome inconsistency if new data from the same gene are added to an analysis already exhibiting a condition indistinguishable from inconsistency. © The Willi Hennig Society 2005.     

A cladistic analysis of the cornutes (stem chordates)

At the root of the calcichordate controversy is the problem of recognizing homologous similarity. Only when a criterion or set of criteria for hypothesizing homologies is established can the calcichordate theory be tested against its main rival—the Stylophoran theory—by applying the principle of parsimony. The criterion of topological similarity is applied to a three-laxon problem involving a mitrate, the crown-group Echinodermata and the crown-group Chordata. The mitrate shares more features with extant chordates than with extant echinoderms and the calcichordate theory is supported by a simple parsimony analysis. In a cladistic analysis of the cornutes, four monophyletic families are recognized (Cothurnocystidae, Scotiaecystidae, Phyllocystidae and Hanusiidae) and their interrelationships resolved using Hennig86 and the successive-weighting procedure of Farris. Because the known fossil record of cornutes and mitrates is very poor, the correlation between the cladogram produced for cornutes and their order of appearance in the geological record is weak. All of the cornute families must have originated in the Lower Cambrian at the latest.     

A cladistic analysis of Geranium subgen. Erodioidea (Picard) Yeo (Geraniaceae)

Subgenus Erodioidea , comprising 19 species, is diagnosed exclusively by the type of fruit discharge and associated morphological characters, all of which are of widespread occurrence within the Geraniaceae. We have no evidence to suggest that the three included sections ( Erodioidea, Aculeolata and Subacaulia) form a monophyletic group. Cladistic analysis of a data set containing 30 morphological characters has confirmed our doubts of monophyly of the subgenus by placing part of the outgroup within the ingroup. Subsequent analyses of the two included non-monotypic sections, Erodioidea and Subacaulia , give different results. For both sections, the monophyly assumption is reasonably well-supported by characters. However, while analysis of sect. Erodioidea results in an explicit hypothesis of relationships and a fully resolved cladogram, resolution within sect. Subacaulia is low. For analysis of the latter, seven different outgroups were used and, in each one, the successive approximations weighting procedure (available in the program Hennig 86) was followed to improve resolution. Despite this, lower nodes in the cladogram remained unresolved, and strict consensus trees of all the analyses resulted in only a few replicated subterminal clades. These results are discussed in connection with biogeographical data and with the hypothesis that representatives of sect. Subacaulia are schizoendemics.     

Plethodontid salamander mitochondrial genomics: A parsimony evaluation of character conflict and implications for historical biogeography

A new parsimony analysis of 27 complete mitochondrial genomic sequences is conducted to investigate the phylogenetic relationships of plethodontid salamanders. This analysis focuses on the amount of character conflict between phylogenetic trees recovered from newly conducted parsimony searches and the Bayesian and maximum likelihood topology reported by Mueller et al. (2004 ; PNAS, 101, 13820–13825). Strong support for Hemidactylium as the sister taxon to all other plethodontids is recovered from parsimony analyses. Plotting area relationships on the most parsimonious phylogenetic tree suggests that eastern North America is the origin of the family Plethodontidae supporting the “Out of Appalachia” hypothesis. A new taxonomy that recognizes clades recovered from phylogenetic analyses is proposed. © The Willi Hennig Society 2005.     

Misleading results of likelihood‐based phylogenetic analyses in the presence of missing data

The amount of missing data in many contemporary phylogenetic analyses has substantially increased relative to previous norms, particularly in supermatrix studies that compile characters from multiple previous analyses. In such cases the missing data are non‐randomly distributed and usually present in all partitions (i.e. groups of characters) sampled. Parametric methods often provide greater resolution and support than parsimony in such cases, yet this may be caused by extrapolation of branch lengths from one partition to another. In this study I use contrived and simulated examples to demonstrate that likelihood, even when applied to simple matrices with little or no homoplasy, homogeneous evolution across groups of characters, perfect model fit, and hundreds or thousands of variable characters, can provide strong support for incorrect topologies when the matrices have non‐random distributions of missing data distributed across all partitions. I do so using a systematic exploration of alternative seven‐taxon tree topologies and distributions of missing data in two partitions to demonstrate that these likelihood‐based artefacts may occur frequently and are not shared by parsimony. I also demonstrate that Bayesian Markov chain Monte Carlo analysis is more robust to these artefacts than is likelihood. © The Willi Hennig Society 2011.     

Quantification of the success of phylogenetic inference in simulations

For phylogenetic simulation studies, the accuracy of topological reconstruction obtained from different data matrices or different methods of phylogenetic inference generally needs to be quantified. Two components of performance within this context are: (1) how the inferred tree topology matches or conflicts with the correct tree topology, and (2) the branch support assigned to both correctly and incorrectly resolved clades. We present a method (averaged overall success of resolution) that incorporates both of these components. Branch support is incorporated in the averaged overall success of resolution by linearly scaling the observed support relative to that conferred by uncontradicted synapomorphies. We believe that this method represents an improvement relative to the commonly used approaches of quantifying the percentage of clades that are correctly resolved in the inferred trees or presenting the Robinson–Foulds distance between the inferred trees and the correct tree. In contrast to Bremer support, the averaged overall success of resolution may be applied equally well to distance, likelihood and parsimony analyses. © The Willi Hennig Society 2006.