Homology as a parsimony problem: a dynamic homology approach for morphological data

The primary data used to reconstruct phylogenies comes organized in the conceptual grid of homology correspondences, and the construction of this theory‐rich grid depends in part on knowledge of relationships. This situation is not satisfactory as a conceptual system, because the evidence is not clearly delimited from the results. I explore the testing of alternative hypotheses of morphological correspondences in a quantitative cladistic context. The varying homology assessments implied by classical criteria of homology (topological equivalence, or position and connections; composition of structures, or commonality in details of construction) can be expressed as regular characters in a cladistic analysis. Doing so provides adequate transformation costs for changes in schemas of correspondences. Correspondences imply evolutionary transformations, and multiple schemas of correspondences can be compared according to the evolutionary transformations that they imply. The method is used to test the correspondences in sclerites of the male copulatory organs of spiders of the subfamily Amaurobioidinae (Arachnida, Araneae, Anyphaenidae). The correspondences of three sclerites are tested, in a data set of 93 species having one, two or three sclerites, using a simultaneous analysis of all the morphological characters. Most parsimonious trees are identified together with the correspondences they imply. Once the correspondences are integrated in the phylogenetic analysis, it is easy to evaluate the robustness of trees or decay in optimality after changes in anatomical interpretations. A Bremer support for anatomical interpretations is proposed, calculated as the increase in tree length when the specific interpretation is not used. © The Willi Hennig Society 2007.     

Homologies in phylogenetic analyses—concept and tests

Analyzing morphological characters in a phylogenetic context comprises two steps, character analysis and cladistic analysis, which are equivalent to two independent tests for hypotheses on homology. The concept of homology concerns comparable parts of the same or different organisms if their correspondences are the consequence of the same genetic or epigenetic information, and consequently of the same origin. The concept of homology is more inclusive than the character concept. Characters are seen as parts of transformation series. In the first step of morphological character analyses correspondences and non-correspondences between two characters are analyzed. A range of different examination methods and accurate study contribute to the severity of test. The hypothesis that two characters are homologous is corroborated if the correspondences outweigh the non-correspondences because the non-correspondences contradict the homology hypothesis whereas the correspondences contradict the analogy hypothesis. Complex characters possess a higher empirical content than less complex characters because they are more severely testable. The cladistic analysis tests characters against other characters which have all passed the first test. Characters which are congruent with the most parsimonious topology are further corroborated; incongruent characters are not seen as ‘falsified’ but as not further corroborated and subject to re-analysis. To test both homologies and topologies repeatedly is consistent with Popperian testability, and it is in such cycles of research that hypotheses will be critically re-evaluated.     

The phylogenetic relationships of lampridiform fishes (Teleostei: acanthomorpha), based on a total-evidence analysis of morphological and molecular data

We investigated the phylogenetic relationships among five species of lampridiform fishes, three basal outgroup species (two aulopiforms and one myctophiform), and two species of non-lampridiform acanthomorphs (Polymixia and Percopsis) using a combined parsimony analysis of morphological and molecular data. Morphological characters included 28 transformation series obtained from the literature. Molecular characters included 223 informative transformation series from an aligned 854-base pair fragment of 12S mtDNA and 139 informative transformation series from an aligned 561-base pair fragment of 16S mtDNA. A total-evidence analysis using the aulopiforms Synodus and Aulopus and the myctophiform Hygophum as outgroups corroborates the monophyly of Lampridiformes and unites Polymixia with Percopsis. Among the lampridiform fishes we examined, Metavelifer is basal, followed in ascending order by Lampris, Lophotus, Regalecus, and Trachipterus. This hypothesis is congruent with the most recent morphological analysis of the Lampridiformes and rejects a diphyletic origin of elongate body form within the clade. Analysis of a combined matrix of 12S and 16S mtDNA data yielded a phylogenetic hypothesis isomorphic with the total-evidence phylogeny. Analyses of partitioned DNA data sets reveals that single gene regions are poor predictors of the total-evidence phylogeny while combined analyses of both DNA data sets are good predictors of the total-evidence phylogeny.     

No speed dating please! Patterns of social preference in male and female house mice

Glass sponges (Class Hexactinellida) are important components of deep-sea ecosystems and are of interest from geological and materials science perspectives. The reconstruction of their phylogeny with molecular data has only recently begun and shows a better agreement with morphology-based systematics than is typical for other sponge groups, likely because of a greater number of informative morphological characters. However, inconsistencies remain that have far-reaching implications for hypotheses about the evolution of their major skeletal construction types (body plans). Furthermore, less than half of all described extant genera have been sampled for molecular systematics, and several taxa important for understanding skeletal evolution are still missing. Increased taxon sampling for molecular phylogenetics of this group is therefore urgently needed. However, due to their remote habitat and often poorly preserved museum material, sequencing all 126 currently recognized extant genera will be difficult to achieve. Utilizing morphological data to incorporate unsequenced taxa into an integrative systematics framework therefore holds great promise, but it is unclear which methodological approach best suits this task. Here, we increase the taxon sampling of four previously established molecular markers (18S, 28S, and 16S ribosomal DNA, as well as cytochrome oxidase subunit I) by 12 genera, for the first time including representatives of the order Aulocalycoida and the type genus of Dactylocalycidae, taxa that are key to understanding hexactinellid body plan evolution. Phylogenetic analyses suggest that Aulocalycoida is diphyletic and provide further support for the paraphyly of order Hexactinosida; hence these orders are abolished from the Linnean classification. We further assembled morphological character matrices to integrate so far unsequenced genera into phylogenetic analyses in maximum parsimony (MP), maximum likelihood (ML), Bayesian, and morphology-based binning frameworks. We find that of these four approaches, total-evidence analysis using MP gave the most plausible results concerning congruence with existing phylogenetic and taxonomic hypotheses, whereas the other methods, especially ML and binning, performed more poorly. We use our total-evidence phylogeny of all extant glass sponge genera for ancestral state reconstruction of morphological characters in MP and ML frameworks, gaining new insights into the evolution of major hexactinellid body plans and other characters such as different spicule types. Our study demonstrates how a comprehensive, albeit in some parts provisional, phylogeny of a larger taxon can be achieved with an integrative approach utilizing molecular and morphological data, and how this can be used as a basis for understanding phenotypic evolution. The datasets and associated trees presented here are intended as a resource and starting point for future work on glass sponge evolution.     

Independence of alignment and tree search

I assert that similarity is the appropriate homology criterion for sequence alignment, as it is with morphology. Methods that select among alignments using parsimony-based tree lengths, as implemented in MALIGN and POY, arrange the data such that they are consistent with a minimum-evolution model. When combining data sets in phylogenetic analyses, we are not trying to reinforce our earlier hypotheses about relationships, but rather to test them. The severity of this test is compromised when congruence with other characters is favored when selecting among alignment parameters.     

A phylogeny of the Platyhelminthes: towards a total-evidence solution

We advocate a total-evidence approach for the reconstruction of working phylogenies for the Turbellaria and the phylum Platyhelminthes. Few morphology-based character matrices are available in the systematic literature concerning flatworms, and molecular-based phylogenies are rapidly providing the only means by which we can estimate phylogenies cladistically. Character matrices based on gross morphology and ultrastructure are required and should be internally consistent, i.e. character coding should follow a set of a priori guidelines and character duplication and contradiction is avoided. In order to test our molecular phylogenies we need complementary data sets from morphology. To understand morphological homology we need phylogenetic evidence from independent (e.g. molecular) data. Fully complementary morphological and molecular data sets enable us to validate phylogenetic hypotheses and the combination of these sets in phylogenetic reconstruction utilises all statements of homology. Working phylogenies which include all phylogenetic information not only shed light on individual character evolution, but form a strong basis for comparative studies investigating the origin and evolutionary radiation of the taxonomic group under scrutiny. This revised version was published online in July 2006 with corrections to the Cover Date.     

On the importance of homology in the age of phylogenomics

Homology is perhaps the most central concept of phylogenetic biology. Molecular systematists have traditionally paid due attention to the homology statements that are implied by their alignments of orthologous sequences, but some authors have suggested that manual gene-by-gene curation is not sustainable in the phylogenomics era. Here, we show that there are multiple ways to efficiently screen for and detect homology errors in phylogenomic data sets. Application of these screening approaches to two phylogenomic data sets, one for birds and another for mammals, shows that these data are replete with homology errors including alignments of different exons to each other, alignments of exons to introns, and alignments of paralogues to each other. The extent of these homology errors weakens the conclusions of studies based on these data sets. Despite advances in automated phylogenomic pipelines, we contend that much of the long, difficult, and sometimes tedious work of systematics is still required to guard against pervasive homology errors. This conclusion is underscored by recent studies that show that just a few outlier genes can impact phylogenetic results at short, tightly spaced internodes that are deep in the Tree of Life. The view that widespread DNA sequence alignment errors are not a major concern for rigorous systematic research is not tenable. If a primary goal of phylogenomics is to resolve the most challenging phylogenetic problems with the abundant data that are now available, researchers must employ effective procedures to screen for and correct homology errors prior to performing downstream phylogenetic analyses.     

Good characters and homology

This paper is a critical comment on a recent article by Lieberman. 1 We question his opinion that DNA is a better data source for phylogenetic reconstructions than bone and discuss his “problems and potential solutions” regarding the homology concept. We conclude that phylogenetic systematics requires a phylogenetic homology concept, and that Lieberman's “solutions,” though useful terms, should not be designated as homology.     

COMBINED AND SEPARATE ANALYSES OF MORPHOLOGICAL AND MOLECULAR DATA IN THE PLANT FAMILY RUBIACEAE

Abstract— The Rubiaceae are one of the largest of the families of angiosperms, with over 10000 species. The tribal and subfamilial classification is provisional due to the lack of phylogenetic hypotheses. The present study of the Rubiaceae is based on 33 genera and three data sets, one morphological and two molecular from chloroplast DNA, restriction sites andrbcL sequences. There is much congruence between the morphological and the molecular data sets, but also conflict. For parsimony reasons, the best phylogenetic hypothesis is a tree based on an analysis of the combined data sets. The so-called “total evidence” criterion for the combined analysis is simply a reiteration of the principle of parsimony. In this particular study, the classification would be almost the same even if based on the separate analyses instead of the combined. Despite the inapplicability of consensus trees or trees from separate analyses for phylogenetic hypotheses and classification, separate analyses may provide important information. It is the best way to reveal conflicts between different data sets. Knowledge of the conflicts can promote further detailed investigation in order to improve understanding of characters and phylogenetic hypotheses. In this study, the tribe Vanguerieae provides such an example; morphological data support a position in the subfamily Cinchonoideae, but DNA and a tree based on the combined data support a position in subfamily Ixoroideae. The tribe's position in the morphological tree is probably due to missing information concerning the correct pollen presentation system. Bootstrap fractions and K. Bremer's branch support values are used to evaluate the stability of particular nodes in the trees. Interestingly these values are not always correlated, e.g. in the morphological tree, the node with the highest branch support value has very low bootstrap fraction. The reasons for these differences are unclear, but large differences are presumably more likely to occur on short branches.     

Total evidence, consensus, and bat phylogeny: A distance-based approach

Resolution of the total evidence (i.e., character congruence) versus consensus (i.e., taxonomic congruence) debate has been impeded by (1) a failure to employ validation methods consistently across both tree-building and consensus analyses, (2) the incomparability of methods for constructing as opposed to those for combining trees, and (3) indifference to aspects of trees other than their topologies. We demonstrate a uniform, distance-based approach which allows for comparability among the results of character- and taxonomic-congruence studies, whether or not an identical suite of taxa has been included in all contributing data sets. Our results indicate that total-evidence and consensus trees differ little in topology if branch lengths are taken into account when combining two or more trees. In addition, when character-state data are converted to distances, our method permits their combination with information produced by techniques which generate distances directly. Moreover, treating all data sets or trees as distance matrices avoids the problem that different numbers of characters in contributing studies may confound the conclusions of a total-evidence or consensus analysis. Our protocol is illustrated with an example involving bats, in which the three component studies based on serology, DNA hybridization, and anatomy imply distinct phylogenies. However, the total-evidence and consensus trees support a fourth, somewhat different, topology resolved at all but one node and which conforms closely to the currently accepted higher category classification of Chiroptera.     

A priori assumptions about characters as a cause of incongruence between molecular and morphological hypotheses of primate interrelationships

When molecules and morphology produce incongruent hypotheses of primate interrelationships, the data are typically viewed as incompatible, and molecular hypotheses are often considered to be better indicators of phylogenetic history. However, it has been demonstrated that the choice of which taxa to include in cladistic analysis as well as assumptions about character weighting, character state transformation order, and outgroup choice all influence hypotheses of relationships and may positively influence tree topology, so that relationships between extant taxa are consistent with those found using molecular data. Thus, the source of incongruence between morphological and molecular trees may lie not in the morphological data themselves but in assumptions surrounding the ways characters evolve and their impact on cladistic analysis. In this study, we investigate the role that assumptions about character polarity and transformation order play in creating incongruence between primate phylogenies based on morphological data and those supported by multiple lines of molecular data. By releasing constraints imposed on published morphological analyses of primates from disparate clades and subjecting those data to parsimony analysis, we test the hypothesis that incongruence between morphology and molecules results from inherent flaws in morphological data. To quantify the difference between incongruent trees, we introduce a new method called branch slide distance (BSD). BSD mitigates many of the limitations attributed to other tree comparison methods, thus allowing for a more accurate measure of topological similarity. We find that releasing a priori constraints on character behavior often produces trees that are consistent with molecular trees. Case studies are presented that illustrate how congruence between molecules and unconstrained morphological data may provide insight into issues of polarity, transformation order, homology, and homoplasy.     

An application of dynamic homology to morphological characters: direct optimization of setae sequences and phylogeny of the family Odontellidae (Poduromorpha, Collembola)

The concept of character and the definition of the attribute are two major theoretical issues of phylogenetics. Lately, great progress has been made in the conceptual development of attributes as historical individuals undergoing series of transformations. While operational application of this ideographic concept of character has been possible since the publication of the direct optimization algorithm and POY software, it has been restricted to molecular characters only. The present paper proposes the first application of direct optimization to morphological characters, in the case study of the phylogeny of Odontellidae. This new homology regime is compared to the traditional homology scheme. The theoretical and operational limitations of the application of direct optimization to morphological characters are discussed. Some thoughts on the basics of its generalization to all morphological characters analyzed in a dynamic homology phylogenetic framework are given.
© The Willi Hennig Society 2009;.     

Assessing similarity: on homology,characters and the need for a semantic approach to non‐evolutionary comparative homology

The problem of homology has been a consistent source of controversy at the heart of systematic biology, as has the step of morphological character analysis in phylogenetics. Based on a clear epistemic framework and a characterization of “characters” as diagnostic evidence units for the recognition of not directly identifiable entities, I discuss the ontological definition and empirical recognition criteria of phylogenetic, developmental and comparative homology, and how these three accounts of homology each contribute to an understanding of the overall phenomenon of homology. I argue that phylogenetic homologies are individuals or historical kinds that require comparative homology for identification. Developmental homologies are natural kinds that ultimately rest on phylogenetic homologies and also require comparative homology for identification. Comparative homologies on the other hand are anatomical structural kinds that are directly identifiable. I discuss pre‐Darwinian comparative homology concepts and their problem of invoking non‐material forces and involving the a priori assumption of a stable positional reference system. Based on Young's concept of comparative homology, I suggest a procedure for recognizing comparative homologues that lacks these problems and that utilizes a semantic framework. This formal conceptual framework provides the much needed semantic transparency and computer‐parsability for documenting, communicating and analysing similarity propositions. It provides an essential methodological framework for generalizing over individual organisms and identifying and demarcating anatomical structural kinds, and it provides the missing link to the logical chain of identifying phylogenetic homology. The approach substantially increases the analytical accessibility of comparative research and thus represents an important contribution to the theoretical and methodological foundation of morphology and comparative biology.     

Phylogenetic analysis of phenotypic characters of Tunicata supports basal Appendicularia and monophyletic Ascidiacea

With approximately 3000 marine species, Tunicata represents the most disparate subtaxon of Chordata. Molecular phylogenetic studies support Tunicata as sister taxon to Craniota, rendering it pivotal to understanding craniate evolution. Although successively more molecular data have become available to resolve internal tunicate phylogenetic relationships, phenotypic data have not been utilized consistently. Herein these shortcomings are addressed by cladistically analyzing 117 phenotypic characters for 49 tunicate species comprising all higher tunicate taxa, and five craniate and cephalochordate outgroup species. In addition, a combined analysis of the phenotypic characters with 18S rDNA-sequence data is performed in 32 OTUs. The analysis of the combined data is congruent with published molecular analyses. Successively up-weighting phenotypic characters indicates that phenotypic data contribute disproportionally more to the resulting phylogenetic hypothesis. The strict consensus tree from the analysis of the phenotypic characters as well as the single most parsimonious tree found in the analysis of the combined dataset recover monophyletic Appendicularia as sister taxon to the remaining tunicate taxa. Thus, both datasets support the hypothesis that the last common ancestor of Tunicata was free-living and that ascidian sessility is a derived trait within Tunicata. “Thaliacea” is found to be paraphyletic with Pyrosomatida as sister taxon to monophyletic Ascidiacea and the relationship between Doliolida and Salpida is unresolved in the analysis of morphological characters; however, the analysis of the combined data reconstructs Thaliacea as monophyletic nested within paraphyletic “Ascidiacea”. Therefore, both datasets differ in the interpretation of the evolution of the complex holoplanktonic life history of thaliacean taxa. According to the phenotypic data, this evolution occurred in the plankton, whereas from the combined dataset a secondary transition into the plankton from a sessile ascidian is inferred. Besides these major differences, both analyses are in accord on many phylogenetic groupings, although both phylogenetic reconstructions invoke a high degree of homoplasy. In conclusion, this study represents the first serious attempt to utilize the potential phylogenetic information present in phenotypic characters to elucidate the inter-relationships of this diverse marine taxon in a consistent cladistic framework.     

Tooth morphology of Echimyidae (Rodentia,Caviomorpha): homology assessments,fossils, and evolution

Echimyidae constitute the most important radiation of caviomorph rodents in the Neotropical region, represented by 20 extant genera and several extinct species. Both in extant and fossil forms, this diversity is reflected by a significant morphological variation found in crown structures of the cheek teeth. Different hypotheses of primary homology have been proposed for these structures, which, in turn, support diverse dental evolutionary hypotheses. In this contribution we inspect the main structures (cusps and lophids) of the lower deciduous teeth and molars in extinct and extant Echimyidae, and establish their topological correspondences. Comparisons with cusps and lophids of Erethizontidae are emphasized. We explore the testing of alternative primary hypotheses of lophid correspondences in a cladistic context. Following a ‘dynamic’ approach, we select the hypothesis of primary homology, which produced the more parsimonious results, and evaluate the evolutionary transformations of the dental characters analysed. In this context, the phylogenetic relationships of living Myocastor coypus (Molina, 1782) with the extinct Tramyocastor and Paramyocastor are tested. Our results indicate that pentalophodonty is the derived condition for the lower molars in Echimyidae, that trilophodonty evolved independently at least three times during the evolutionary history of these rodents, and that tetralophodonty represents the plesiomorphic condition. This study shows that dental evolution in echimyids can be better understood when occlusal structures are expressed as reliably comparable characters, and when fossils are taken into account. © 2012 The Linnean Society of London, Zoological Journal of the Linnean Society, 2012, 164 , 451–480.     

SuperFine: fast and accurate supertree estimation

Many research groups are estimating trees containing anywhere from a few thousands to hundreds of thousands of species, toward the eventual goal of the estimation of a Tree of Life, containing perhaps as many as several million leaves. These phylogenetic estimations present enormous computational challenges, and current computational methods are likely to fail to run even on data sets in the low end of this range. One approach to estimate a large species tree is to use phylogenetic estimation methods (such as maximum likelihood) on a supermatrix produced by concatenating multiple sequence alignments for a collection of markers; however, the most accurate of these phylogenetic estimation methods are extremely computationally intensive for data sets with more than a few thousand sequences. Supertree methods, which assemble phylogenetic trees from a collection of trees on subsets of the taxa, are important tools for phylogeny estimation where phylogenetic analyses based upon maximum likelihood (ML) are infeasible. In this paper, we introduce SuperFine, a meta-method that utilizes a novel two-step procedure in order to improve the accuracy and scalability of supertree methods. Our study, using both simulated and empirical data, shows that SuperFine-boosted supertree methods produce more accurate trees than standard supertree methods, and run quickly on very large data sets with thousands of sequences. Furthermore, SuperFine-boosted matrix representation with parsimony (MRP, the most well-known supertree method) approaches the accuracy of ML methods on supermatrix data sets under realistic conditions.     

A phylogenetic analysis of Coelopidae (Diptera) based on morphological and DNA sequence data

The phylogenetic relationships of 22 species of Coelopidae are reconstructed based on a data matrix consisting of morphological and DNA sequence characters (16S rDNA, EF-1alpha). Optimal gap and transversion costs are determined via a sensitivity analysis and both equal weighting and a transversion cost of 2 are found to perform best based on taxonomic congruence, character incongruence, and tree support. The preferred phylogenetic hypothesis is fully resolved and well-supported by jackknife, bootstrap, and Bremer support values, but it is in conflict with the cladogram based on morphological characters alone. Most notably, the Coelopidae and the genus Coelopa are not monophyletic. However, partitioned Bremer Support and an analysis of node stability under different gap and transversion costs reveal that the critical clades rendering these taxa non-monophyletic are poorly supported. Furthermore, the monophyly of Coelopidae and Coelopa is not rejected in analyses using 16S rDNA that was manually aligned. The resolution of the tree based on this reduced data sets is, however, lower than for the tree based on the full data sets. Partitioned Bremer support values reveal that 16S rDNA characters provide the largest amount of tree support, but the support values are heavily dependent on analysis conditions. Problems with direct comparison of branch support values for trees derived using fixed alignments with those obtained under optimization alignment are discussed. Biogeographic history and available behavioral and genetic data are also discussed in light of this first cladogram for Coelopidae based on a quantitative phylogenetic analysis.     

More characters or more taxa for a robust phylogeny--case study from the coffee family (Rubiaceae)

Using different data sets mainly from the plant family Rubiaceae, but in parts also from the Apocynaceae, Asteraceae, Lardizabalaceae, Saxifragaceae, and Solanaceae, we have investigated the effect of number of characters, number of taxa, and kind of data on bootstrap values within phylogenetic trees. The percentage of supported nodes within a tree is positively correlated with the number of characters, and negatively correlated with the number of taxa. The morphological analyses are based on few characters and weakly supported trees are expected. The percentage of supported nodes is also dependent on the kind of data analyzed. In analyses of Rubiaceae based on the same number of characters, RFLP data give trees with higher percentage of supported nodes than rbcL and morphological data. We also discuss the support values for particular nodes at the familial and subfamilial levels. Two new data sets of ndhF and rbcL sequences of Rubiaceae are analyzed and together with earlier studies of the family we can conclude that the monophyly of the Rubiaceae is supported and within the family there are three well supported, but not easily characterized, large subfamilies, Rubioideae, Cinchonoideae s.s. and Ixoroideae s.l. There are also a few genera (Luculia and Coptosapelta) unclassified to subfamily.     

Avian higher-level phylogeny: well-supported clades and what we can learn from a phylogenetic analysis of 2954 morphological characters

It has been shown that increased character sampling betters the accuracy of phylogenetic reconstructions in the case of molecular data. A recently published analysis of avian higher-level phylogenetics based on 2954 morphological characters now provides an empirical example to test whether this is also true in the case of morphological characters. Several clades are discussed which are supported by multiple analyses of mutually independent molecular data (sequences of nuclear genes on different chromosomes and mitochondrial genes) as well as morphological apomorphies, but did not result from parsimony analysis of the large morphological data set. Incorrect character scorings in that analysis notwithstanding, it is concluded that in the case of morphological data, increased character sampling does not necessarily better the accuracy of a phylogenetic reconstruction. Because morphological characters usually have a strongly varying complexity, many simple and homoplastic characters may overrule fewer ones of greater phylogenetic significance in large data sets, thus producing a low ratio of phylogenetic signal to 'noise' in the data.