Euclidean nature of phylogenetic distance matrices

Phylogenies are fundamental to comparative biology as they help to identify independent events on which statistical tests rely. Two groups of phylogenetic comparative methods (PCMs) can be distinguished: those that take phylogenies into account by introducing explicit models of evolution and those that only consider phylogenies as a statistical constraint and aim at partitioning trait values into a phylogenetic component (phylogenetic inertia) and one or multiple specific components related to adaptive evolution. The way phylogenetic information is incorporated into the PCMs depends on the method used. For the first group of methods, phylogenies are converted into variance-covariance matrices of traits following a given model of evolution such as Brownian motion (BM). For the second group of methods, phylogenies are converted into distance matrices that are subsequently transformed into Euclidean distances to perform principal coordinate analyses. Here, we show that simply taking the elementwise square root of a distance matrix extracted from a phylogenetic tree ensures having a Euclidean distance matrix. This is true for any type of distances between species (patristic or nodal) and also for trees harboring multifurcating nodes. Moreover, we illustrate that this simple transformation using the square root imposes less geometric distortion than more complex transformations classically used in the literature such as the Cailliez method. Given the Euclidean nature of the elementwise square root of phylogenetic distance matrices, the positive semidefinitiveness of the phylogenetic variance-covariance matrix of a trait following a BM model, or related models of trait evolution, can be established. In that way, we build a bridge between the two groups of statistical methods widely used in comparative analysis. These results should be of great interest for ecologists and evolutionary biologists performing statistical analyses incorporating phylogenies.     

Inferring protein interactions from phylogenetic distance matrices

Finding the interacting pairs of proteins between two different protein families whose members are known to interact is an important problem in molecular biology. We developed and tested an algorithm that finds optimal matches between two families of proteins by comparing their distance matrices. A distance matrix provides a measure of the sequence similarity of proteins within a family. Since the protein sets of interest may have dozens of proteins each, the use of an efficient approximate solution is necessary. Therefore the approach we have developed consists of a Metropolis Monte Carlo optimization algorithm which explores the search space of possible matches between two distance matrices. We demonstrate that by using this algorithm we are able to accurately match chemokines and chemokine-receptors as well as the tgfbeta family of ligands and their receptors.     

Assessing DNA sequence variations in human ESTs in a phylogenetic context using high-density oligonucleotide arrays

We have analyzed human genomic diversity in 32 individuals representing four continental populations of Homo sapiens in the context of four ape species. We used DNA resequencing chips covering 898 expressed sequence tags (ESTs), corresponding to 109 kb of sequence. Based on the intra-species data, the neutral hypothesis could not be rejected. However, the mutation rate was two times lower than typically observed in functionally unconstrained genomic segments, suggesting a certain level of selection. The worldwide diversity (297 segregating sites and nucleotide diversity of 0.054%) was partitioned among continents, with the greatest amount of variation observed in the African sample. The long-term effective population size of the human population was estimated at 13,000; a similar figure was obtained for the African sample and a 20% lower estimate was obtained for the other continents. Africans also differed in having a higher number of continental-specific polymorphisms contributing to the higher average nucleotide diversity. These results are consistent with the existence of two distinct lineages of modern humans: amalgamation of these lineages in Africa led to the higher present-day diversity on that continent, whereas colonization of other continents by one of them gave the effect of a population bottleneck.     

Testing for phylogenetic signal in phenotypic traits: new matrices of phylogenetic proximities

Abouheif adapted a test for serial independence to detect a phylogenetic signal in phenotypic traits. We provide the exact analytic value of this test, revealing that it uses Moran's I statistic with a new matrix of phylogenetic proximities. We introduce then two new matrices of phylogenetic proximities highlighting their mathematical properties: matrix A which is used in Abouheif test and matrix M which is related to A and biodiversity studies. Matrix A unifies the tests developed by Abouheif, Moran and Geary. We discuss the advantages of matrices A and M over three widely used phylogenetic proximity matrices through simulations evaluating power and type-I error of tests for phylogenetic autocorrelation. We conclude that A enhances the power of Moran's test and is useful for unresolved trees. Data sets and routines are freely available in an online package and explained in an online supplementary file.     

Assessing site-interdependent phylogenetic models of sequence evolution

In recent works, methods have been proposed for applying phylogenetic models that allow for a general interdependence between the amino acid positions of a protein. As of yet, such models have focused on site interdependencies resulting from sequence-structure compatibility constraints, using simplified structural representations in combination with a set of statistical potentials. This structural compatibility criterion is meant as a proxy for sequence fitness, and the methods developed thus far can incorporate different site-interdependent fitness proxies based on other measurements. However, no methods have been proposed for comparing and evaluating the adequacy of alternative fitness proxies in this context, or for more general comparisons with canonical models of protein evolution. In the present work, we apply Bayesian methods of model selection-based on numerical calculations of marginal likelihoods and posterior predictive checks-to evaluate models encompassing the site-interdependent framework. Our application of these methods indicates that considering site-interdependencies, as done here, leads to an improved model fit for all data sets studied. Yet, we find that the use of pairwise contact potentials alone does not suitably account for across-site rate heterogeneity or amino acid exchange propensities; for such complexities, site-independent treatments are still called for. The most favored models combine the use of statistical potentials with a suitably rich site-independent model. Altogether, the methodology employed here should allow for a more rigorous and systematic exploration of different ways of modeling explicit structural constraints, or any other site-interdependent criterion, while best exploiting the richness of previously proposed models.     

Mitochondrial DNA coding region sequences support the phylogenetic distinction of two Indian wolf species

Two small endangered populations of Indian wolves were recently shown to be distant from other wolf and dog mtDNA lineages characterized so far. None of the inner branches in the tree of canid species based on partial hypervariable D-loop sequences were, however, statistically supported by the data raising the question whether the two Indian wolf lineages represent two new species, occupying an intermediate position between Canis latrans and C. lupus or have diverged from the sub-species of C. lupus due to isolation and drift. Here we report complete D-loop, cytochrome b, and 16S rRNA sequences data for 23 additional wolves from India analysed in the context of other canid species. Extended analyses of D-loop data and partial sequences of 16S rRNA showed highly reticulated pattern and were unable to resolve unambiguously the phylogenetic relationship of Indian wolves among other canid species. The phylogenetic reconstructions of cytochrome b sequences, however gave significant statistical support for the inner branches supporting genetic distinction of the two Indian wolf lineages within themselves as well as from all other wolves of the world, including individuals belonging to subspecies C. lupus chanco and C. lupus pallipes to which the two Indian wolf populations have been traditionally assigned. Their genetic differentiation relative to worldwide variation of wolves supports the suggestion to treat them as separate wolf species, C. himalayensis and C. indica .     

Assessing the functional turnover of species assemblages with tailored dissimilarity matrices

It is often suggested that community functional diversity is an appropriate predictive measure of ecosystem functioning, particularly if relevant species traits for the ecological property of interest are carefully selected. However, methods for selecting traits are often based on expert knowledge or on theoretical models of ecosystem functioning, but usually do not include explicitly developed quantitative procedures. Here we propose to construct a so‐called ‘tailored dissimilarity matrix’ between species assemblages to emphasize their functional turnover in response to some user‐defined ecological property. First, a subset of community weighted mean trait values (CWM) is selected by stepwise regression on the ecological process of interest. The selected CWM values are then replaced by the residuals of the least‐squares regressions of each single CWM on the ecological process of interest and pairwise Euclidean distances between the residual values at each sampling site are calculated. We illustrate the advantages of the tailored approach using two distinct plant and bee communities under contrasting fire regimes in temperate forests of southern Switzerland. Our results demonstrated that, unlike for the original CWM values, the tailored approach optimized the degree of functional differentiation among bee and plant species assemblages, i.e. the species functional turnover, with respect to different fire regimes.     

Molecular evolution and phylogenetic implications of internal transcribed spacer sequences of Ribosomal DNA in Winteraceae

The internal transcribed spacers and the 5.8S coding region of nuclear ribosomal DNA were sequenced and analyzed to address questions of generic relationships in Winteraceae. The molecular data generated a single tree that is congruent with one based on morphological data. The sequences of ITS 1 in the family range from 235 to 252 bases in size and of ITS 2 from 213 to 226 bases. The size of the 5.8S coding region is 164 bases. The range of ITS 1 and ITS 2 sequence divergence between pairs of genera within Winteraceae is relatively low in comparison to other plant families. Two types of ITS 1 and ITS 2 sequences were observed in the same individual for some taxa. Sequence variations between the two arrays are 4.7%–6.3% for ITS 1 and 5.1%–7.0% for ITS 2. Both arrays of sequences, however, generate the same phylogenetic relationships. Rates of nucleotide substitutions for the internal transcribed spacers are 3.2–5.2 × 10^-10 substitution per site per year estimated in ITS 1 and 3.6–5.7 × 10^-10 in ITS 2.     

Fast phylogenetic DNA barcoding

We present a heuristic approach to the DNA assignment problem based on phylogenetic inferences using constrained neighbour joining and non-parametric bootstrapping. We show that this method performs as well as the more computationally intensive full Bayesian approach in an analysis of 500 insect DNA sequences obtained from GenBank. We also analyse a previously published dataset of environmental DNA sequences from soil from New Zealand and Siberia, and use these data to illustrate the fact that statistical approaches to the DNA assignment problem allow for more appropriate criteria for determining the taxonomic level at which a particular DNA sequence can be assigned.     

Assessing the passerine "Tapestry": phylogenetic relationships of the Muscicapoidea inferred from nuclear DNA sequences

This study presents new comparative sequence data from the nuclear RAG-1 gene for an increased taxon sample in order to investigate phylogenetic relationships among a diverse songbird superfamily, the Muscicapoidea, which has variously included the waxwings, silky flycatchers, Palm Chat, dippers, starlings, mockingbirds, thrushes, chats, and Old World flycatchers. At the same time, our results provide a test of the often-cited relationships inferred from the phenetic studies of Sibley and Ahlquist [Phylogeny and Classification of Birds: A Study in Molecular Evolution. Yale University Press, New Haven, 1990] using DNA hybridization distances. Nuclear DNA sequences confirm the monophyly of the "core muscicapoid" group, as defined by Barker et al. [Proc. R. Soc. Lond. B 269 (2002) 295] and also support the sister-group relationship of the Sturnidae and Mimidae, on the one hand, and the large-bodied thrushes (Turdini)+the Old World flycatchers and robins, on the other. The results of the phylogenetic analysis allow preliminary inferences about muscicapoid biogeographic history.     

The challenge of constructing large phylogenetic trees

The amount of sequence data available to reconstruct the evolutionary history of genes and species has increased 20-fold in the past decade. Consequently the size of phylogenetic analyses has grown as well, and phylogenetic methods, algorithms and their implementations have struggled to keep pace. Computational and other challenges raised by this burgeoning database emerge at several stages of analysis, from the optimal assembly of large data matrices from sequence databases, to the efficient construction of trees from these large matrices and the piece-wise assembly of 'supertrees' from those trees in turn. A final challenge is posed by the difficulty of visualizing and making inferences from trees that might soon routinely contain thousands of species.     

Estimating divergence times in large phylogenetic trees

A new method, PATHd8, for estimating ultrametric trees from trees with edge (branch) lengths proportional to the number of substitutions is proposed. The method allows for an arbitrary number of reference nodes for time calibration, each defined either as absolute age, minimum age, or maximum age, and the tree need not be fully resolved. The method is based on estimating node ages by mean path lengths from the node to the leaves but correcting for deviations from a molecular clock suggested by reference nodes. As opposed to most existing methods allowing substitution rate variation, the new method smoothes substitution rates locally, rather than simultaneously over the whole tree, thus allowing for analysis of very large trees. The performance of PATHd8 is compared with other frequently used methods for estimating divergence times. In analyses of three separate data sets, PATHd8 gives similar divergence times to other methods, the largest difference being between crown group ages, where unconstrained nodes get younger ages when analyzed with PATHd8. Overall, chronograms obtained from other methods appear smoother, whereas PATHd8 preserves more of the heterogeneity seen in the original edge lengths. Divergence times are most evenly spread over the chronograms obtained from the Bayesian implementation and the clock-based Langley-Fitch method, and these two methods produce very similar ages for most nodes. Evaluations of PATHd8 using simulated data suggest that PATHd8 is slightly less precise compared with penalized likelihood, but it gives more sensible answers for extreme data sets. A clear advantage with PATHd8 is that it is more or less instantaneous even with trees having several thousand leaves, whereas other programs often run into problems when analyzing trees with hundreds of leaves. PATHd8 is implemented in freely available software.     

Assessing ancient DNA studies

The study of ancient DNA has the potential to make significant and unique contributions to ecology and evolution. However, the techniques used contain inherent problems, particularly with regards to the generation of authentic and useful data. The solution currently advocated to reduce contamination and artefactual results is to adopt criteria for authentication. Nevertheless, these criteria are not foolproof, and we believe that they have, in practice, replaced the use of thought and prudence when designing and executing ancient DNA studies. We argue here that researchers in this field must take a more cognitive and self-critical approach. Specifically, in place of checking criteria off lists, researchers must explain, in sufficient enough detail to dispel doubt, how the data were obtained, and why they should be believed to be authentic.     

我国赫坎按蚊复合体成员种的rDNA-ITS2区序列差异及系统发育分析

测定了我国赫坎按蚊复合体 9成员种的核糖体DNA第二内转录间隔区 (rDNA ITS2 )序列 ,根据序列差异分析各蚊种间的系统发育关系。结果显示 :( 1 )ITS2区序列最长的是中华按蚊 ( 4 6 8bp) ,最短的是克劳按蚊和赫坎按蚊 ( 4 36bp) ;GC含量为 4 4 9%～ 4 6 8% ;( 2 )发现该复合体 4成员种的ITS2区序列存在种内个体间差异 ,幅度为 0～ 3 8% ,明显小于种间差异 ;( 3)将各蚊种的ITS2区序列进行同源排序比较 ,发现其变异大多是简单重复单元的拷贝数不同 ;种间差异性最大的是克劳按蚊与嗜人按蚊( 32 3% ) ,最小的是贵阳按蚊与凉山按蚊 ( 9 0 % )平均差异率为 2 2 3% ;( 4 )根据ITS2区序列特征 ,用 3种方法构建的树状图拟合一致。以上结果表明赫坎按蚊复合体各成员种rDNA ITS2序列在种内非常保守 ,以种间序列差异分析为基础的分子鉴别技术是甄别蚊种分类地位混淆和错误的有效方法。