2013
DOI: 10.1093/molbev/mst030
|View full text |Cite
|
Sign up to set email alerts
|

FUBAR: A Fast, Unconstrained Bayesian AppRoximation for Inferring Selection

Abstract: Model-based analyses of natural selection often categorize sites into a relatively small number of site classes. Forcing each site to belong to one of these classes places unrealistic constraints on the distribution of selection parameters, which can result in misleading inference due to model misspecification. We present an approximate hierarchical Bayesian method using a Markov chain Monte Carlo (MCMC) routine that ensures robustness against model misspecification by averaging over a large number of predefin… Show more

Help me understand this report

Search citation statements

Order By: Relevance

Paper Sections

Select...
2
1
1
1

Citation Types

8
971
0

Year Published

2014
2014
2024
2024

Publication Types

Select...
9

Relationship

0
9

Authors

Journals

citations
Cited by 1,118 publications
(1,024 citation statements)
references
References 40 publications
8
971
0
Order By: Relevance
“…The methods used to investigate positive codon sites included FEL, SLAC, FUBAR, MEME,62, 63, 64 the branch site REL and the GA‐branch site models were chosen to determine the selection pressure on the individual branches 65, 66. Methods with p < 0.1 in SLAC, p < 0.05 in FEL and MEME and the posterior probability >0.9 in FUBAR, were considered to be more conservative positive selection pressure.…”
Section: Methodsmentioning
confidence: 99%
“…The methods used to investigate positive codon sites included FEL, SLAC, FUBAR, MEME,62, 63, 64 the branch site REL and the GA‐branch site models were chosen to determine the selection pressure on the individual branches 65, 66. Methods with p < 0.1 in SLAC, p < 0.05 in FEL and MEME and the posterior probability >0.9 in FUBAR, were considered to be more conservative positive selection pressure.…”
Section: Methodsmentioning
confidence: 99%
“…We further implemented likelihood and Bayesian-based methods to identify site-specific Cytb positive selection where the rate of non-synonymous substitution (dN) is greater than the rate of synonymous substitution (dS). We applied a suit of different algorithms for selection inference to our data: single likelihood ancestor counting (SLAC), fixed effects likelihood (FEL) [30], internal fixed effects likelihood (IFEL) [31], fast unconstrained Bayesian approximation (FUBAR) [32] and mixed effects model of evolution (MEME) [33] to our data. SLAC is based on the reconstruction of the ancestral sequences and the counts of dS and dN at each codon position of the phylogeny.…”
Section: Methodsmentioning
confidence: 99%
“…In addition to CODEML models, we used the programme HyPhy implemented in the web-server Datamonkey (Delport et al, 2010) to identify codons putatively under selection using Fast Unconstrained Bayesian AppRoximation (FUBAR; Murrell et al, 2013) and Mixed Effects Model of Evolution (MEME; Murrell et al, 2012). FUBAR assigns each codon a posterior probability (PP) of belonging to three classes of ω: ω 0 o1, ω 1 = 1 and ω 0 41.…”
Section: Divergence Time Estimatesmentioning
confidence: 99%