The Behavior of Metropolis-Coupled Markov Chains When Sampling Rugged Phylogenetic Distributions

被引:4
作者
Brown, Jeremy M. [1 ,2 ]
Thomson, Robert C. [3 ]
机构
[1] Louisiana State Univ, Dept Biol Sci, 202 Life Sci Bldg, Baton Rouge, LA 70803 USA
[2] Louisiana State Univ, Museum Nat Sci, 202 Life Sci Bldg, Baton Rouge, LA 70803 USA
[3] Univ Hawaii, Dept Biol, 2538 McCarthy Mall,Edmondson Hall Room 216, Honolulu, HI 96822 USA
基金
美国国家科学基金会;
关键词
Metropolis coupling; Markov chain Monte Carlo; Bayesian phylogenetic inference; INFERENCE;
D O I
10.1093/sysbio/syy008
中图分类号
Q [生物科学];
学科分类号
07 ; 0710 ; 09 ;
摘要
Bayesian phylogenetic inference relies on the use of Markov chain Monte Carlo (MCMC) to provide numerical approximations of high-dimensional integrals and estimate posterior probabilities. However, MCMC performs poorly when posteriors are very rugged (i.e., regions of high posterior density are separated by regions of low posterior density). One technique that has become popular for improving numerical estimates from MCMC when distributions are rugged is Metropolis coupling (MC3). InMC3, additional chains are employed to sample flattened transformations of the posterior and improve mixing. Here, we highlight several underappreciated behaviors of MC3. Notably, estimated posterior probabilities may be incorrect but appear to converge, when individual chains do not mix well, despite different chains sampling trees from all relevant areas in tree space. Counterintuitively, such behavior can be more difficult to diagnose with increased numbers of chains. We illustrate these surprising behaviors of MC3 using a simple, non-phylogenetic example and phylogenetic examples involving both constrained and unconstrained analyses. To detect and mitigate the effects of these behaviors, we recommend increasing the number of independent analyses and varying the temperature of the hottest chain in current versions of Bayesian phylogenetic software. Convergence diagnostics based on the behavior of the hottest chain may also help detect these behaviors and could form a useful addition to future software releases.
引用
收藏
页码:729 / 734
页数:6
相关论文
共 9 条
  • [1] Parallel metropolis coupled Markov chain Monte Carlo for Bayesian phylogenetic inference
    Altekar, G
    Dwarkadas, S
    Huelsenbeck, JP
    Ronquist, F
    [J]. BIOINFORMATICS, 2004, 20 (03) : 407 - 415
  • [2] Bayesian Tests of Topology Hypotheses with an Example from Diving Beetles
    Bergsten, Johannes
    Nilsson, Anders N.
    Ronquist, Fredrik
    [J]. SYSTEMATIC BIOLOGY, 2013, 62 (05) : 660 - 673
  • [3] Bayes Factors Unmask Highly Variable Information Content, Bias, and Extreme Influence in Phylogenomic Analyses
    Brown, Jeremy M.
    Thomson, Robert C.
    [J]. SYSTEMATIC BIOLOGY, 2017, 66 (04) : 517 - 530
  • [4] More than 1000 ultraconserved elements provide evidence that turtles are the sister group of archosaurs
    Crawford, Nicholas G.
    Faircloth, Brant C.
    McCormack, John E.
    Brumfield, Robb T.
    Winker, Kevin
    Glenn, Travis C.
    [J]. BIOLOGY LETTERS, 2012, 8 (05) : 783 - 786
  • [5] Geyer C.J., 1991, Computing Science and Statistics, P153
  • [6] Three crocodilian genomes reveal ancestral patterns of evolution among archosaurs
    Green, Richard E.
    Braun, Edward L.
    Armstrong, Joel
    Earl, Dent
    Nguyen, Ngan
    Hickey, Glenn
    Vandewege, Michael W.
    St John, John A.
    Capella-Gutierrez, Salvador
    Castoe, Todd A.
    Kern, Colin
    Fujita, Matthew K.
    Opazo, Juan C.
    Jurka, Jerzy
    Kojima, Kenji K.
    Caballero, Juan
    Hubley, Robert M.
    Smit, Arian F.
    Platt, Roy N.
    Lavoie, Christine A.
    Ramakodi, Meganathan P.
    Finger, John W.
    Suh, Alexander
    Isberg, Sally R.
    Miles, Lee
    Chong, Amanda Y.
    Jaratlerdsiri, Weerachai
    Gongora, Jaime
    Moran, Christopher
    Iriarte, Andres
    McCormack, John
    Burgess, Shane C.
    Edwards, Scott V.
    Lyons, Eric
    Williams, Christina
    Breen, Matthew
    Howard, Jason T.
    Gresham, Cathy R.
    Peterson, Daniel G.
    Schmitz, Juergen
    Pollock, David D.
    Haussler, David
    Triplett, Eric W.
    Zhang, Guojie
    Irie, Naoki
    Jarvis, Erich D.
    Brochu, Christopher A.
    Schmidt, Carl J.
    McCarthy, Fiona M.
    Faircloth, Brant C.
    [J]. SCIENCE, 2014, 346 (6215) : 1335 - +
  • [7] MrBayes 3: Bayesian phylogenetic inference under mixed models
    Ronquist, F
    Huelsenbeck, JP
    [J]. BIOINFORMATICS, 2003, 19 (12) : 1572 - 1574
  • [8] MrBayes 3.2: Efficient Bayesian Phylogenetic Inference and Model Choice Across a Large Model Space
    Ronquist, Fredrik
    Teslenko, Maxim
    van der Mark, Paul
    Ayres, Daniel L.
    Darling, Aaron
    Hohna, Sebastian
    Larget, Bret
    Liu, Liang
    Suchard, Marc A.
    Huelsenbeck, John P.
    [J]. SYSTEMATIC BIOLOGY, 2012, 61 (03) : 539 - 542
  • [9] Yang Z, 2014, Molecular Evolution: A Statistical Approach, P1, DOI 10.1093/acprof:oso/9780199602605.001.0001