Improving marginal likelihood estimation for Bayesian phylogenetic model selection.

By W. Xie, Paul O. Lewis, Yu Fan, Lynn Kuo, Ming-Hui Chen. Systematic Biology, 2011

995 citations recorded by the source corpus.

Subfields: Genetics, Molecular Biology, Paleontology

View Paper: https://www.semanticscholar.org/paper/8b29a58ed29bfcdc7eeb7d3da21a05ea22871442

How researchers see this paper

The researchers whose trusted networks rank it highest, strongest advocate first. Every researcher gets a different order.

In Genetics

In Molecular Biology

In Paleontology

Citations in the corpus

The Thermodynamic Variational Objective (2019, methodology): “TI is now commonly used in phylogenetics to calculate the Bayes factor B = p ( x | M 1 ) /p ( x | M 0 ) , where M 0 , M 1 are two models specifying (for instance) tree topologies and branch lengths (Lartillot and Philippe, 2006; Xie et al., 2010; Rodrigue and Aris-Brosou, 2011).”
Territoriality, Social Bonds, and the Evolution of Communal Signaling in Birds (2016, positive): “For each chain, the marginal likelihood was calculated using a stepping stone sampler (Xie et al., 2011): 100 stones were distributed according to a beta distribution (shape 0.400000, scale 1.000000) and each stone was run for 25,000 iterations.”
Estimating the Marginal Likelihood Using the Arithmetic Mean Identity (2017, methodology): “Using the method presented in Xie et al. (2011), we have showed that even if P (A) is known, the expected value of the Corrected Harmonic Mean times P (A) is greater than p(y), i.e. p(y) < Eθ|y ⎛ ⎝P (A) [ 1 k k∑ q=1 IA(θ(q)) p(y|θ(q)) ]−1⎞⎠ (10) (see Supplementary Appendix A in Pajor (2016)).”
Estimating the timing of early eukaryotic diversification with multigene molecular clocks (2011, methodology): “If competing models converged in likelihood scores, the likelihoods were compared using Bayes factors (37) as assessed in Tracer, although we did not rely on this metric as the harmonic mean calculation of Bayes factors has been demonstrated to be unreliable (38).”
Bayesian phylogenetic and phylodynamic data integration using BEAST 1.10 (2018, positive): “…likelihood estimation (Baele et al. 2012), using path sampling (Gelman and Meng 1998; Lartillot and Philippe 2006) and stepping-stone sampling (Xie et al. 2011), as well as the recently developed generalized stepping-stone sampling (Fan et al. 2011; Baele et al. 2016a) that offers increased…”
Improving the accuracy of demographic and molecular clock model comparison while accommodating phylogenetic uncertainty. (2012, methodology): “Xie et al. (2011) propose to calculate the marginal likelihood using n samples from a series of K + 1 power posteriors as follows: (5) The authors show that numerical stability can be improved by factoring out the largest sampled likelihood for each power posterior.”
Evaluating marginal likelihood with thermodynamic integration method and comparison with several other numerical methods (2016, methodology): “There have been other MC methods for evaluating the marginal likelihood, such as those using SavageDickey density ratio [Morey et al., 2011], Lebesgue integration theory [Weinberg, 2012], and stepping-stone method [Xie et al., 2011].”
A novel Bayesian method for inferring and interpreting the dynamics of adaptive landscapes from phylogenetic comparative data (2014, methodology): “Each power posterior is used as an importance density to estimate a series of ratios of normalizing constants, the product of which provides the ratio of the known reference distribution and the marginal likelihood (Xie et al. 2011).”
A Total-Evidence Approach to Dating with Fossils, Applied to the Early Radiation of the Hymenoptera (2012, methodology): “They were calculated from estimates of the marginal likelihoods obtained using the stepping stone sampling approach (Xie et al. 2011), which we implemented in MrBayes 3.2.”
Improving power posterior estimation of statistical evidence (2012, methodology): “(17) If this intersecting point is outside the interval [tk, tk+1], it suggests there is some sort of inflection within the interval and instead we use a simple weighted average.”
Accurate model selection of relaxed molecular clocks in bayesian phylogenetics. (2012, methodology): “We calculate marginal likelihoods using PS (Gelman and Meng 1998; Ogata 1989; Lartillot and Philippe 2006), SS (Xie et al. 2011), HME (Newton and Raftery 1994), and sHME (Redelings and Suchard 2005).”
MrBayes 3.2: Efficient Bayesian Phylogenetic Inference and Model Choice Across a Large Model Space (2012, methodology): “Of these, MrBayes 3.2 implements the recently proposed stepping stone method (Xie et al. 2011) that uses MCMC to sample from a series of so-called power posterior distributions connecting the posterior distribution with the prior distribution.”
The early spread and epidemic ignition of HIV-1 in human populations (2014, methodology): “Marginal likelihoods of parametric models (constant, expansion, exponential, logistic growth), and a recently developed flexible nonparametric model (90), were compared on data set C using an improved variant of path sampling (PS) (91, 92) called stepping-stone (SS) (93) sampling.”