Comparing Bootstrap and Posterior Probability Values in the Four-Taxon Case

Abstract

Assessment of the reliability of a given phylogenetic hypothesis is an important step in phylogenetic analysis. Historically, the nonparametric bootstrap procedure has been the most frequently used method for assessing the support for specific phylogenetic relationships. The recent employment of Bayesian methods for phylogenetic inference problems has resulted in clade support being expressed in terms of posterior probabilities. We used simulated data and the four-taxon case to explore the relationship between nonparametric bootstrap values (as inferred by maximum likelihood) and posterior probabilities (as inferred by Bayesian analysis). The results suggest a complex association between the two measures. Three general regions of tree space can be identified: (1) the neutral zone, where differences between mean bootstrap and mean posterior probability values are not significant, (2) near the two-branch corner, and (3) deep in the two-branch corner. In the last two regions, significant differences occur between mean bootstrap and mean posterior probability values. Whether bootstrap or posterior probability values are higher depends on the data in support of alternative topologies. Examination of star topologies revealed that both bootstrap and posterior probability values differ significantly from theoretical expectations; in particular, there are more posterior probability values in the range 0.85-1 than expected by theory. Therefore, our results corroborate the findings of others that posterior probability values are excessively high. Our results also suggest that extrapolations from single topology branch-length studies are unlikely to provide any general conclusions regarding the relationship between bootstrap and posterior probability values. [Bayesian analysis; Markov chain Monte Carlo sampling; maximum likelihood; phylogenetics.] Posterior ProbabilitiesAn important development in phylogenetic analysis has been the application of Bayesian methods, particularly when coupled with Markov chain Monte Carlo (MCMC) methods (

Molecular phylogenetic analyses reveal the importance of taxon sampling in cryptic diversity: Liolaemus nigroviridis and L. monticola (Liolaeminae) as focal species

“…The ML topology shows two major groups, one supported by maximum likelihood bootstrap (MLB = 92) and by Bayesian posterior probabilities (BPP = 0.99; clade A), and a second clade not supported by maximum likelihood bootstrap (MLB = 47) but supported by Bayesian posterior probabilities (BPP = 0.96; clade B). The difference in the support for the clade B may result from a variety of causes, including the different nature of both analyses and overestimation of posterior probabilities [52][53][54][55]. The two sequences of L. uniformis were placed in a supported clade (MLB = 100, BPP = 1.0) within the L. nigroviridis group, related (but unsupported) to the supported clade that included samples from La Campana, El Roble, and Chicauma (Fig.…”

Polyphyly of Polycarpon (Caryophyllaceae) inferred from DNA sequence data

“…Of the recovered clades some are only supported by PP values. It has already been shown that Bayesian statistics can overestimate support for particular nodes on a tree (Cummings & al., 2003;Simmons & al., 2004). On the other hand, it has been reported that Bayesian analysis can provide high support values for correct clades with fewer characters than needed for bootstrap (Alfaro & al., 2003).…”

The genome of the protist parasite Entamoeba histolytica

“…A consensus tree was made from the remaining samples. Because posterior probabilities-the support values used by bayesian analysis to indicate confidence in groups-have been criticized 29 , we also used bootstrapping to provide an additional indication of support for relationships. Each data set was bootstrapped (100 replicates) and used to make distance matrices under the same evolutionary model as in the bayesian analysis, using custom (P4) software (available on request).…”