Survey of Branch Support Methods Demonstrates Accuracy, Power, and Robustness of Fast Likelihood-based Approximation Schemes

Abstract

Phylogenetic inference and evaluating support for inferred relationships is at the core of many studies testing evolutionary hypotheses. Despite the popularity of nonparametric bootstrap frequencies and Bayesian posterior probabilities, the interpretation of these measures of tree branch support remains a source of discussion. Furthermore, both methods are computationally expensive and become prohibitive for large data sets. Recent fast approximate likelihood-based measures of branch supports (approximate likelihood ratio test [aLRT] and Shimodaira–Hasegawa [SH]-aLRT) provide a compelling alternative to these slower conventional methods, offering not only speed advantages but also excellent levels of accuracy and power. Here we propose an additional method: a Bayesian-like transformation of aLRT (aBayes). Considering both probabilistic and frequentist frameworks, we compare the performance of the three fast likelihood-based methods with the standard bootstrap (SBS), the Bayesian approach, and the recently introduced rapid bootstrap. Our simulations and real data analyses show that with moderate model violations, all tests are sufficiently accurate, but aLRT and aBayes offer the highest statistical power and are very fast. With severe model violations aLRT, aBayes and Bayesian posteriors can produce elevated false-positive rates. With data sets for which such violation can be detected, we recommend using SH-aLRT, the nonparametric version of aLRT based on a procedure similar to the Shimodaira–Hasegawa tree selection. In general, the SBS seems to be excessively conservative and is much slower than our approximate likelihood-based methods.

Rock comminution as a source of hydrogen for subglacial ecosystems

“…Water-rock chemolithotrophic reactions could provide additional sources of energy to sustain subglacial ecosystems 11 . Abiogenic H2, formed by a variety of water-rock reactions including serpentinization 12,13 , radiolysis 14 or mineral surface radical reactions 15 , provides an energy substrate for a variety of warm to hot (≥22°C) subsurface environments.…”

A new species of tapir from the Amazon

“…After a full PhyML search on tree topology and parameters, the maximumlikelihood tree was used again as an initial tree for a new run and the search was repeated until no significant gain in likelihood was observed in a likelihood-ratio test. The confidence in the clades was assessed through the approximate likelihood-ratio (aLRT) test (Anisimova and Gascuel 2006;Anisimova et al 2011). Second, we used a Bayesian inference method with MrBayes 3.2.1 (Ronquist et al 2011) using a GTRþCþinv model of nucleotide evolution and assuming flat Dirichlet priors for parameter estimates and unconstrained branch lengths.…”

Description of a new species of Homonota (Reptilia, Squamata, Phyllodactylidae) from the central region of northern Paraguay

“…These analyses were made for each gene individually and for a concatenated dataset of the three genes together. For the ML analysis we used IQ-Tree (Nguyen et al 2015) through its webserver (Trifinopoulos et al 2016) using 10,000 non parametric bootstrap replicates plus 10,000 replicates of Shimodaira-Hasegawa approximate likelihood ratio (SH-aLRT) (Anisimova et al 2011) and 10,000 ultrafast bootstrap (UFBoot) approximation replicates (Minh et al 2013). We converted the alignment to nexus format in the online server Alter (Glez-Peña et al 2010) available at http://sing.ei.uvigo.es/ALTER/, to be used in MrBayes v3.2 Ronquist 2001, Ronquist andHuelsenbeck 2003) for a BI.…”