Diego Darriba

jModelTest 2: more models, new heuristics and parallel computing

jModelTest 2: more models, new heuristics and parallel computing Diego Darriba, Guillermo L. Taboada, Ramón Doallo and David Posada Supplementary Table 1. New features in jModelTest 2 Supplementary Table 2. Model selection accuracy Supplementary Table 3. Mean square errors for model averaged estimates Supplementary Note 1. Hill-climbing hierarchical clustering algorithm Supplementary Note 2. Heuristic filtering Supplementary Note 3. Simulations from prior distributions Supplementary Note 4. Speed-up benchmark on real and simulated datasets

ProtTest 3

UNLABELLED We have implemented a high-performance computing (HPC) version of ProtTest that can be executed in parallel in multicore desktops and clusters. This version, called ProtTest 3, includes new features and extended capabilities. AVAILABILITY ProtTest 3 source code and binaries are freely available under GNU license for download from http://darwin.uvigo.es/software/prottest3, linked to a Mercurial repository at Bitbucket (https://bitbucket.org/). CONTACT dposada@uvigo.es SUPPLEMENTARY INFORMATION Supplementary data are available at Bioinformatics online.

ProtTest-HPC: fast selection of best-fit models of protein evolution

The use of probabilistic models of amino acid replacement is essential for the study of protein evolution, and programs like ProtTest implement different strategies to identify the best-fit model for the data at hand. For large protein alignments, this task can demand vast computational resources, preventing the justification of the model used in the analysis. We have implemented a High Performance Computing (HPC) version of ProtTest. ProtTest-HPC can be executed in parallel in HPC environments as: (1) a GUI-based desktop version that uses multi-core processors and (2) a cluster-based version that distributes the computational load among nodes. The use of ProtTest-HPC resulted in significant performance gains, with speedups of up to 50 on a high performance cluster.

The phylogenetic likelihood library.

We introduce the Phylogenetic Likelihood Library (PLL), a highly optimized application programming interface for developing likelihood-based phylogenetic inference and postanalysis software. The PLL implements appropriate data structures and functions that allow users to quickly implement common, error-prone, and labor-intensive tasks, such as likelihood calculations, model parameter as well as branch length optimization, and tree space exploration. The highly optimized and parallelized implementation of the phylogenetic likelihood function and a thorough documentation provide a framework for rapid development of scalable parallel phylogenetic software. By example of two likelihood-based phylogenetic codes we show that the PLL improves the sequential performance of current software by a factor of 2–10 while requiring only 1 month of programming time for integration. We show that, when numerical scaling for preventing floating point underflow is enabled, the double precision likelihood calculations in the PLL are up to 1.9 times faster than those in BEAGLE. On an empirical DNA dataset with 2000 taxa the AVX version of PLL is 4 times faster than BEAGLE (scaling enabled and required). The PLL is available at http://www.libpll.org under the GNU General Public License (GPL).

jmodeltest.org: selection of nucleotide substitution models on the cloud

Summary: The selection of models of nucleotide substitution is one of the major steps of modern phylogenetic analysis. Different tools exist to accomplish this task, among which jModelTest 2 (jMT2) is one of the most popular. Still, to deal with large DNA alignments with hundreds or thousands of loci, users of jMT2 need to have access to High Performance Computing clusters, including installation and configuration capabilities, conditions not always met. Here we present jmodeltest.org, a novel web server for the transparent execution of jMT2 across different platforms and for a wide range of users. Its main benefit is straightforward execution, avoiding any configuration/execution issues, and reducing significantly in most cases the time required to complete the analysis. Availability and implementation: jmodeltest.org is accessible using modern browsers, such as Firefox, Chrome, Opera, Safari and IE from http://jmodeltest.org. User registration is not mandatory, but users wanting to have additional functionalities, like access to previous analyses, have the possibility of opening a user account. Contact: info@jmodeltest.org

Does the choice of nucleotide substitution models matter topologically

BackgroundIn the context of a master level programming practical at the computer science department of the Karlsruhe Institute of Technology, we developed and make available an open-source code for testing all 203 possible nucleotide substitution models in the Maximum Likelihood (ML) setting under the common Akaike, corrected Akaike, and Bayesian information criteria. We address the question if model selection matters topologically, that is, if conducting ML inferences under the optimal, instead of a standard General Time Reversible model, yields different tree topologies. We also assess, to which degree models selected and trees inferred under the three standard criteria (AIC, AICc, BIC) differ. Finally, we assess if the definition of the sample size (#sites versus #sites × #taxa) yields different models and, as a consequence, different tree topologies.ResultsWe find that, all three factors (by order of impact: nucleotide model selection, information criterion used, sample size definition) can yield topologically substantially different final tree topologies (topological difference exceeding 10 %) for approximately 5 % of the tree inferences conducted on the 39 empirical datasets used in our study.ConclusionsWe find that, using the best-fit nucleotide substitution model may change the final ML tree topology compared to an inference under a default GTR model. The effect is less pronounced when comparing distinct information criteria. Nonetheless, in some cases we did obtain substantial topological differences.

The State of Software for Evolutionary Biology

Abstract With Next Generation Sequencing data being routinely used, evolutionary biology is transforming into a computational science. Thus, researchers have to rely on a growing number of increasingly complex software. All widely used core tools in the field have grown considerably, in terms of the number of features as well as lines of code and consequently, also with respect to software complexity. A topic that has received little attention is the software engineering quality of widely used core analysis tools. Software developers appear to rarely assess the quality of their code, and this can have potential negative consequences for end‐users. To this end, we assessed the code quality of 16 highly cited and compute‐intensive tools mainly written in C/C++ (e.g., MrBayes, MAFFT, SweepFinder, etc.) and JAVA (BEAST) from the broader area of evolutionary biology that are being routinely used in current data analysis pipelines. Because, the software engineering quality of the tools we analyzed is rather unsatisfying, we provide a list of best practices for improving the quality of existing tools and list techniques that can be deployed for developing reliable, high quality scientific software from scratch. Finally, we also discuss journal as well as science policy and, more importantly, funding issues that need to be addressed for improving software engineering quality as well as ensuring support for developing new and maintaining existing software. Our intention is to raise the awareness of the community regarding software engineering quality issues and to emphasize the substantial lack of funding for scientific software development.

The impact of partitioning on phylogenomic accuracy

Several strategies have been proposed to assign substitution models in phylogenomic datasets, or partitioning. The accuracy of these methods, and most importantly, their impact on phylogenetic estimation has not been thoroughly assessed using computer simulations. We simulated multiple partitioning scenarios to benchmark two a priori partitioning schemes (one model for the whole alignment, one model for each data block), and two statistical approaches (hierarchical clustering and greedy) implemented in PartitionFinder and in our new program, PartitionTest. Most methods were able to identify optimal partitioning schemes closely related to the true one. Greedy algorithms identified the true partitioning scheme more frequently than the clustering algorithms, but selected slightly less accurate partitioning schemes and tended to underestimate the number of partitions. PartitionTest was several times faster than PartitionFinder, with equal or better accuracy. Importantly, maximum likelihood phylogenetic inference was very robust to the partitioning scheme. Best-fit partitioning schemes resulted in optimal phylogenetic performance, without appreciable differences compared to the use of the true partitioning scheme. However, accurate trees were also obtained by a “simple” strategy consisting of assigning independent GTR+G models to each data block. On the contrary, leaving the data unpartitioned always diminished the quality of the trees inferred, to a greater or lesser extent depending on the simulated scenario. The analysis of empirical data confirmed these trends, although suggesting a stronger influence of the partitioning scheme. Overall, our results suggests that statistical partitioning, but also the a priori assignment of independent GTR+G models, maximize phylogenomic performance.

High-performance computing selection of models of DNA substitution for multicore clusters

This paper presents the high-performance computing (HPC) support of jModelTest2, the most popular bioinformatic tool for the statistical selection of models of DNA substitution. As this can demand vast computational resources, especially in terms of processing power, jModelTest2 implements three parallel algorithms for model selection: (1) a multithreaded implementation for shared memory architectures; (2) a message-passing implementation for distributed memory architectures, such as clusters; and (3) a hybrid shared/distributed memory implementation for clusters of multicore nodes, combining the workload distribution across cluster nodes with a multithreaded model optimization within each node. The main limitation of the shared and distributed versions is the workload imbalance that generally appears when using more than 32 cores, a direct consequence of the heterogeneity in the computational cost of the evaluated models. The hybrid shared/distributed memory version overcomes this issue reducing the workload imbalance through a thread-based decomposition of the most costly model optimization tasks. The performance evaluation of this HPC application on a 40-core shared memory system and on a 528-core cluster has shown high scalability, with speedups of the multithreaded version of up to 32, and up to 257 for the hybrid shared/distributed memory implementation. This can represent a reduction in the execution time of some analyses from 4 days down to barely 20 minutes. The implementation of the three parallel execution strategies of jModelTest2 presented in this paper are available under a GPL license at http://code.google.com/jmodeltest2.

HPC selection of models of DNA substitution

Statistical model selection has become an essential step for the estimation of phylogenies from DNA sequence alignments. The program jModelTest offers different strategies to identify best-fit models for the data at hand, but for large DNA alignments, this task can demand vast computational resources. This paper presents a High Performance Computing (HPC) adaptation of jModelTest for shared memory multi-core systems and distributed memory cluster platforms. The performance evaluation of this HPC version on a shared memory system and on a cluster shows significant performance advantages, with speedups up to 39. This could represent a reduction in the execution time of some analyses from almost one day to half an hour.