Alexandros Stamatakis

RAxML-III: a fast program for maximum likelihood-based inference of large phylogenetic trees

MOTIVATION The computation of large phylogenetic trees with statistical models such as maximum likelihood or bayesian inference is computationally extremely intensive. It has repeatedly been demonstrated that these models are able to recover the true tree or a tree which is topologically closer to the true tree more frequently than less elaborate methods such as parsimony or neighbor joining. Due to the combinatorial and computational complexity the size of trees which can be computed on a Biologists PC workstation within reasonable time is limited to trees containing approximately 100 taxa. RESULTS In this paper we present the latest release of our program RAxML-III for rapid maximum likelihood-based inference of large evolutionary trees which allows for computation of 1.000-taxon trees in less than 24 hours on a single PC processor. We compare RAxML-III to the currently fastest implementations for maximum likelihood and bayesian inference: PHYML and MrBayes. Whereas RAxML-III performs worse than PHYML and MrBayes on synthetic data it clearly outperforms both programs on all real data alignments used in terms of speed and final likelihood values. Availability SUPPLEMENTARY INFORMATION RAxML-III including all alignments and final trees mentioned in this paper is freely available as open source code at http://wwwbode.cs.tum/~stamatak CONTACT stamatak@cs.tum.edu.

Whole-genome analyses resolve early branches in the tree of life of modern birds

To better determine the history of modern birds, we performed a genome-scale phylogenetic analysis of 48 species representing all orders of Neoaves using phylogenomic methods created to handle genome-scale data. We recovered a highly resolved tree that confirms previously controversial sister or close relationships. We identified the first divergence in Neoaves, two groups we named Passerea and Columbea, representing independent lineages of diverse and convergently evolved land and water bird species. Among Passerea, we infer the common ancestor of core landbirds to have been an apex predator and confirm independent gains of vocal learning. Among Columbea, we identify pigeons and flamingoes as belonging to sister clades. Even with whole genomes, some of the earliest branches in Neoaves proved challenging to resolve, which was best explained by massive protein-coding sequence convergence and high levels of incomplete lineage sorting that occurred during a rapid radiation after the Cretaceous-Paleogene mass extinction event about 66 million years ago.

Assessing the root of bilaterian animals with scalable phylogenomic methods

A clear picture of animal relationships is a prerequisite to understand how the morphological and ecological diversity of animals evolved over time. Among others, the placement of the acoelomorph flatworms, Acoela and Nemertodermatida, has fundamental implications for the origin and evolution of various animal organ systems. Their position, however, has been inconsistent in phylogenetic studies using one or several genes. Furthermore, Acoela has been among the least stable taxa in recent animal phylogenomic analyses, which simultaneously examine many genes from many species, while Nemertodermatida has not been sampled in any phylogenomic study. New sequence data are presented here from organisms targeted for their instability or lack of representation in prior analyses, and are analysed in combination with other publicly available data. We also designed new automated explicit methods for identifying and selecting common genes across different species, and developed highly optimized supercomputing tools to reconstruct relationships from gene sequences. The results of the work corroborate several recently established findings about animal relationships and provide new support for the placement of other groups. These new data and methods strongly uphold previous suggestions that Acoelomorpha is sister clade to all other bilaterian animals, find diminishing evidence for the placement of the enigmatic Xenoturbella within Deuterostomia, and place Cycliophora with Entoprocta and Ectoprocta. The work highlights the implications that these arrangements have for metazoan evolution and permits a clearer picture of ancestral morphologies and life histories in the deep past.

How many bootstrap replicates are necessary

Phylogenetic bootstrapping (BS) is a standard technique for inferring confidence values on phylogenetic trees that is based on reconstructing many trees from minor variations of the input data, trees called replicates. BS is used with all phylogenetic reconstruction approaches, but we focus here on one of the most popular, maximum likelihood (ML). Because ML inference is so computationally demanding, it has proved too expensive to date to assess the impact of the number of replicates used in BS on the relative accuracy of the support values. For the same reason, a rather small number (typically 100) of BS replicates are computed in real-world studies. Stamatakis et al. recently introduced a BS algorithm that is 1 to 2 orders of magnitude faster than previous techniques, while yielding qualitatively comparable support values, making an experimental study possible. In this article, we propose stopping criteria--that is, thresholds computed at runtime to determine when enough replicates have been generated--and we report on the first large-scale experimental study to assess the effect of the number of replicates on the quality of support values, including the performance of our proposed criteria. We run our tests on 17 diverse real-world DNA--single-gene as well as multi-gene--datasets, which include 125-2,554 taxa. We find that our stopping criteria typically stop computations after 100-500 replicates (although the most conservative criterion may continue for several thousand replicates) while producing support values that correlate at better than 99.5% with the reference values on the best ML trees. Significantly, we also find that the stopping criteria can recommend very different numbers of replicates for different datasets of comparable sizes. Our results are thus twofold: (i) they give the first experimental assessment of the effect of the number of BS replicates on the quality of support values returned through BS, and (ii) they validate our proposals for stopping criteria. Practitioners will no longer have to enter a guess nor worry about the quality of support values; moreover, with most counts of replicates in the 100-500 range, robust BS under ML inference becomes computationally practical for most datasets. The complete test suite is available at http://lcbb.epfl.ch/BS.tar.bz2, and BS with our stopping criteria is included in the latest release of RAxML v7.2.5, available at http://wwwkramer.in.tum.de/exelixis/software.html.

Phylogenetic models of rate heterogeneity: a high performance computing perspective

Inference of phylogenetic trees using the maximum likelihood (ML) method is NP-hard. Furthermore, the computation of the likelihood function for huge trees of more than 1,000 organisms is computationally intensive due to a large amount of floating point operations and high memory consumption. Within this context, the present paper compares two competing mathematical models that account for evolutionary rate heterogeneity: the Gamma and CAT models. The intention of this paper is to show that - from a purely empirical point of view - CAT can be used instead of Gamma. The main advantage of CAT over Gamma consists in significantly lower memory consumption and faster inference times. An experimental study using RAxML has been performed on 19 real-world datasets comprising 73 up to 1,663 DNA sequences. Results show that CAT is on average 5.5 times faster than Gamma and - surprisingly enough - also yields trees with slightly superior Gamma likelihood values. The usage of the CAT model decreases the amount of average L2 and L3 cache misses by factor 8.55

Selecting optimal partitioning schemes for phylogenomic datasets

BackgroundPartitioning involves estimating independent models of molecular evolution for different subsets of sites in a sequence alignment, and has been shown to improve phylogenetic inference. Current methods for estimating best-fit partitioning schemes, however, are only computationally feasible with datasets of fewer than 100 loci. This is a problem because datasets with thousands of loci are increasingly common in phylogenetics.MethodsWe develop two novel methods for estimating best-fit partitioning schemes on large phylogenomic datasets: strict and relaxed hierarchical clustering. These methods use information from the underlying data to cluster together similar subsets of sites in an alignment, and build on clustering approaches that have been proposed elsewhere.ResultsWe compare the performance of our methods to each other, and to existing methods for selecting partitioning schemes. We demonstrate that while strict hierarchical clustering has the best computational efficiency on very large datasets, relaxed hierarchical clustering provides scalable efficiency and returns dramatically better partitioning schemes as assessed by common criteria such as AICc and BIC scores.ConclusionsThese two methods provide the best current approaches to inferring partitioning schemes for very large datasets. We provide free open-source implementations of the methods in the PartitionFinder software. We hope that the use of these methods will help to improve the inferences made from large phylogenomic datasets.

How Many Bootstrap Replicates Are Necessary

Phylogenetic Bootstrapping (BS) is a standard technique for inferring confidence values on phylogenetic trees that is based on reconstructing many trees from minor variations of the input data, trees called replicates. BS is used with all phylogenetic reconstruction approaches, but we focus here on the most popular, Maximum Likelihood (ML). Because ML inference is so computationally demanding, it has proved too expensive to date to assess the impact of the number of replicates used in BS on the quality of the support values. For the same reason, a rather small number (typically 100) of BS replicates are computed in real-world studies. Stamatakis et al. recently introduced a BS algorithm that is 1---2 orders of magnitude faster than previous techniques, while yielding qualitatively comparable support values, making an experimental study possible. In this paper, we propose stopping criteria , that is, thresholds computed at runtime to determine when enough replicates have been generated, and report on the first large-scale experimental study to assess the effect of the number of replicates on the quality of support values, including the performance of our proposed criteria. We run our tests on 17 diverse real-world DNA, single-gene as well as multi-gene, datasets, that include between 125 and 2,554 sequences. We find that our stopping criteria typically stop computations after 100---500 replicates (although the most conservative criterion may continue for several thousand replicates) while producing support values that correlate at better than 99.5% with the reference values on the best ML trees. Significantly, we also find that the stopping criteria can recommend very different numbers of replicates for different datasets of comparable sizes. Our results are thus two-fold: (i) they give the first experimental assessment of the effect of the number of BS replicates on the quality of support values returned through bootstrapping; and (ii) they validate our proposals for stopping criteria. Practitioners will no longer have to enter a guess nor worry about the quality of support values; moreover, with most counts of replicates in the 100---500 range, robust BS under ML inference becomes computationally practical for most datasets. The complete test suite is available at http://lcbb.epfl.ch/BS.tar.bz2 and BS with our stopping criteria is included in RAxML 7.1.0.

Metagenomic species profiling using universal phylogenetic marker genes

To quantify known and unknown microorganisms at species-level resolution using shotgun sequencing data, we developed a method that establishes metagenomic operational taxonomic units (mOTUs) based on single-copy phylogenetic marker genes. Applied to 252 human fecal samples, the method revealed that on average 43% of the species abundance and 58% of the richness cannot be captured by current reference genome–based methods. An implementation of the method is available at http://www.bork.embl.de/software/mOTU/.

Large-scale maximum likelihood-based phylogenetic analysis on the IBM BlueGene/L

Phylogenetic inference is a grand challenge in Bioinformatics due to immense computational requirements. The increasing popularity of multi-gene alignments in biological studies, which typically provide a stable topological signal due to a more favorable ratio of the number of base pairs to the number of sequences, coupled with rapid accumulation of sequence data in general, poses new challenges for high performance computing. In this paper, we demonstrate how state-of-the-art Maximum Likelihood (ML) programs can be efficiently scaled to the IBM BlueGene/L (BG/L) architecture, by porting RAxML, which is currently among the fastest and most accurate programs for phylogenetic inference under the ML criterion. We simultaneously exploit coarse-grained and fine-grained parallelism that is inherent in every ML-based biological analysis. Performance is assessed using datasets consisting of 212 sequences and 566,470 base pairs, and 2,182 sequences and 51,089 base pairs, respectively. To the best of our knowledge, these are the largest datasets analyzed under ML to date. The capability to analyze such datasets will help to address novel biological questions via phylogenetic analyses. Our experimental results indicate that the fine-grained parallelization scales well up to 1, 024 processors. Moreover, a larger number of processors can be efficiently exploited by a combination of coarse-grained and fine-grained parallelism. Finally, we demonstrate that our parallelization scales equally well on an AMD Opteron cluster with a less favorable network latency to processor speed ratio. We recorded super-linear speedups in several cases due to increased cache efficiency.

ExaBayes: Massively Parallel Bayesian Tree Inference for the Whole-Genome Era

Modern sequencing technology now allows biologists to collect the entirety of molecular evidence for reconstructing evolutionary trees. We introduce a novel, user-friendly software package engineered for conducting state-of-the-art Bayesian tree inferences on data sets of arbitrary size. Our software introduces a nonblocking parallelization of Metropolis-coupled chains, modifications for efficient analyses of data sets comprising thousands of partitions and memory saving techniques. We report on first experiences with Bayesian inferences at the whole-genome level using the SuperMUC supercomputer and simulated data.