Relaxed molecular clocks, the bias-variance trade-off, and the quality of phylogenetic inference

Joel O. Wertheim, Michael Sanderson, Michael Worobey, Adam Bjork

Research output: Contribution to journalArticle

46 Citations (Scopus)

Abstract

Because a constant rate of DNA sequence evolution cannot be assumed to be ubiquitous, relaxed molecular clock inference models have proven useful when estimating rates and divergence dates. Furthermore, it has been recently suggested that using relaxed molecular clocks may provide superior accuracy and precision in phylogenetic inference compared with traditional time-free methods that do not incorporate a molecular clock. We perform a simulation study to determine if assuming a relaxed molecular clock does indeed improve the quality of phylogenetic inference. We analyze sequence data simulated under various rate distributions using relaxed-clocks, strict-clocks, and time-free Bayesian phylogenetic inference models. Our results indicate that no difference exists in the quality of phylogenetic inference between assuming a relaxed molecular clock and making no assumption about the clock-likeness of sequence evolution. This pattern is likely due to the bias-variance trade-off inherent in this type of phylogenetic inference. We also compared the quality of inference between Bayesian and maximum likelihood time-free inference models and found them to be qualitatively similar.

Original languageEnglish (US)
Pages (from-to)1-8
Number of pages8
JournalSystematic Biology
Volume59
Issue number1
DOIs
StatePublished - Jan 2010

Fingerprint

trade-off
phylogenetics
phylogeny
Sequence Analysis
divergence
nucleotide sequences
DNA
simulation
rate
methodology

Keywords

  • Bayesian
  • Bias-variance trade-off
  • Maximum likelihood
  • Relaxed molecular clock
  • Robinson-Foulds tree-to-tree distance

ASJC Scopus subject areas

  • Ecology, Evolution, Behavior and Systematics
  • Genetics
  • Medicine(all)

Cite this

Relaxed molecular clocks, the bias-variance trade-off, and the quality of phylogenetic inference. / Wertheim, Joel O.; Sanderson, Michael; Worobey, Michael; Bjork, Adam.

In: Systematic Biology, Vol. 59, No. 1, 01.2010, p. 1-8.

Research output: Contribution to journalArticle

@article{bad951ec31ca40c6a1ac92b823b5b2b7,
title = "Relaxed molecular clocks, the bias-variance trade-off, and the quality of phylogenetic inference",
abstract = "Because a constant rate of DNA sequence evolution cannot be assumed to be ubiquitous, relaxed molecular clock inference models have proven useful when estimating rates and divergence dates. Furthermore, it has been recently suggested that using relaxed molecular clocks may provide superior accuracy and precision in phylogenetic inference compared with traditional time-free methods that do not incorporate a molecular clock. We perform a simulation study to determine if assuming a relaxed molecular clock does indeed improve the quality of phylogenetic inference. We analyze sequence data simulated under various rate distributions using relaxed-clocks, strict-clocks, and time-free Bayesian phylogenetic inference models. Our results indicate that no difference exists in the quality of phylogenetic inference between assuming a relaxed molecular clock and making no assumption about the clock-likeness of sequence evolution. This pattern is likely due to the bias-variance trade-off inherent in this type of phylogenetic inference. We also compared the quality of inference between Bayesian and maximum likelihood time-free inference models and found them to be qualitatively similar.",
keywords = "Bayesian, Bias-variance trade-off, Maximum likelihood, Relaxed molecular clock, Robinson-Foulds tree-to-tree distance",
author = "Wertheim, {Joel O.} and Michael Sanderson and Michael Worobey and Adam Bjork",
year = "2010",
month = "1",
doi = "10.1093/sysbio/syp072",
language = "English (US)",
volume = "59",
pages = "1--8",
journal = "Systematic Biology",
issn = "1063-5157",
publisher = "Oxford University Press",
number = "1",

}

TY - JOUR

T1 - Relaxed molecular clocks, the bias-variance trade-off, and the quality of phylogenetic inference

AU - Wertheim, Joel O.

AU - Sanderson, Michael

AU - Worobey, Michael

AU - Bjork, Adam

PY - 2010/1

Y1 - 2010/1

N2 - Because a constant rate of DNA sequence evolution cannot be assumed to be ubiquitous, relaxed molecular clock inference models have proven useful when estimating rates and divergence dates. Furthermore, it has been recently suggested that using relaxed molecular clocks may provide superior accuracy and precision in phylogenetic inference compared with traditional time-free methods that do not incorporate a molecular clock. We perform a simulation study to determine if assuming a relaxed molecular clock does indeed improve the quality of phylogenetic inference. We analyze sequence data simulated under various rate distributions using relaxed-clocks, strict-clocks, and time-free Bayesian phylogenetic inference models. Our results indicate that no difference exists in the quality of phylogenetic inference between assuming a relaxed molecular clock and making no assumption about the clock-likeness of sequence evolution. This pattern is likely due to the bias-variance trade-off inherent in this type of phylogenetic inference. We also compared the quality of inference between Bayesian and maximum likelihood time-free inference models and found them to be qualitatively similar.

AB - Because a constant rate of DNA sequence evolution cannot be assumed to be ubiquitous, relaxed molecular clock inference models have proven useful when estimating rates and divergence dates. Furthermore, it has been recently suggested that using relaxed molecular clocks may provide superior accuracy and precision in phylogenetic inference compared with traditional time-free methods that do not incorporate a molecular clock. We perform a simulation study to determine if assuming a relaxed molecular clock does indeed improve the quality of phylogenetic inference. We analyze sequence data simulated under various rate distributions using relaxed-clocks, strict-clocks, and time-free Bayesian phylogenetic inference models. Our results indicate that no difference exists in the quality of phylogenetic inference between assuming a relaxed molecular clock and making no assumption about the clock-likeness of sequence evolution. This pattern is likely due to the bias-variance trade-off inherent in this type of phylogenetic inference. We also compared the quality of inference between Bayesian and maximum likelihood time-free inference models and found them to be qualitatively similar.

KW - Bayesian

KW - Bias-variance trade-off

KW - Maximum likelihood

KW - Relaxed molecular clock

KW - Robinson-Foulds tree-to-tree distance

UR - http://www.scopus.com/inward/record.url?scp=72049088557&partnerID=8YFLogxK

UR - http://www.scopus.com/inward/citedby.url?scp=72049088557&partnerID=8YFLogxK

U2 - 10.1093/sysbio/syp072

DO - 10.1093/sysbio/syp072

M3 - Article

C2 - 20525616

AN - SCOPUS:72049088557

VL - 59

SP - 1

EP - 8

JO - Systematic Biology

JF - Systematic Biology

SN - 1063-5157

IS - 1

ER -