Graphs in phylogenetic comparative analysis: Anscombe's quartet revisited

Liam J. Revell, Klaus Schliep, Eugenio Valderrama, James E. Richardson

Resultado de la investigación: Contribución a RevistaArtículo

1 Cita (Scopus)

Resumen

In 1973, the statistician Francis Anscombe used a clever set of bivariate datasets (now known as Anscombe's quartet) to illustrate the importance of graphing data as a component of statistical analyses. In his example, each of the four datasets yielded identical regression coefficients and model fits, and yet when visualized revealed strikingly different patterns of covariation between x and y. Phylogenetic comparative methods (the set of methodologies that use phylogenies, often combined with phenotypic trait data, to make inferences about evolution) are statistical methods too; yet visualizing the data and phylogeny in a sensible way that would permit us to detect unexpected patterns or unanticipated deviations from model assumptions is not a routine component of phylogenetic comparative analyses. Here, we use a quartet of phylogenetic datasets to illustrate that the same estimated parameters and model fits can be obtained from data that were generated using markedly different procedures—including pure Brownian motion evolution and randomly selected data uncorrelated with the tree. Just as in the case of Anscombe's quartet, when graphed the differences between the four datasets are quickly revealed. The intent of this article is to help build the general case that phylogenetic comparative methods are statistical methods and consequently that graphing or visualization should invariably be included as an essential step in our standard data analytical pipelines. Phylogenies are complex data structures and thus visualizing data on trees in a meaningful and useful way is a challenging endeavour. We recommend that the development of graphical methods for simultaneously visualizing data and tree should continue to be an important goal in phylogenetic comparative biology.

Idioma originalEnglish (US)
Páginas (desde-hasta)2145-2154
Número de páginas10
PublicaciónMethods in Ecology and Evolution
Volumen9
N.º10
DOI
EstadoPublished - oct 1 2018

All Science Journal Classification (ASJC) codes

  • Ecology, Evolution, Behavior and Systematics
  • Ecological Modeling

Citar esto

Revell, Liam J. ; Schliep, Klaus ; Valderrama, Eugenio ; Richardson, James E. / Graphs in phylogenetic comparative analysis : Anscombe's quartet revisited. En: Methods in Ecology and Evolution. 2018 ; Vol. 9, N.º 10. pp. 2145-2154.
@article{3bddcf146c674f1ca55691eae05049bc,
title = "Graphs in phylogenetic comparative analysis: Anscombe's quartet revisited",
abstract = "In 1973, the statistician Francis Anscombe used a clever set of bivariate datasets (now known as Anscombe's quartet) to illustrate the importance of graphing data as a component of statistical analyses. In his example, each of the four datasets yielded identical regression coefficients and model fits, and yet when visualized revealed strikingly different patterns of covariation between x and y. Phylogenetic comparative methods (the set of methodologies that use phylogenies, often combined with phenotypic trait data, to make inferences about evolution) are statistical methods too; yet visualizing the data and phylogeny in a sensible way that would permit us to detect unexpected patterns or unanticipated deviations from model assumptions is not a routine component of phylogenetic comparative analyses. Here, we use a quartet of phylogenetic datasets to illustrate that the same estimated parameters and model fits can be obtained from data that were generated using markedly different procedures—including pure Brownian motion evolution and randomly selected data uncorrelated with the tree. Just as in the case of Anscombe's quartet, when graphed the differences between the four datasets are quickly revealed. The intent of this article is to help build the general case that phylogenetic comparative methods are statistical methods and consequently that graphing or visualization should invariably be included as an essential step in our standard data analytical pipelines. Phylogenies are complex data structures and thus visualizing data on trees in a meaningful and useful way is a challenging endeavour. We recommend that the development of graphical methods for simultaneously visualizing data and tree should continue to be an important goal in phylogenetic comparative biology.",
author = "Revell, {Liam J.} and Klaus Schliep and Eugenio Valderrama and Richardson, {James E.}",
year = "2018",
month = "10",
day = "1",
doi = "10.1111/2041-210X.13067",
language = "English (US)",
volume = "9",
pages = "2145--2154",
journal = "Methods in Ecology and Evolution",
issn = "2041-210X",
publisher = "British Ecological Society",
number = "10",

}

Graphs in phylogenetic comparative analysis : Anscombe's quartet revisited. / Revell, Liam J.; Schliep, Klaus; Valderrama, Eugenio; Richardson, James E.

En: Methods in Ecology and Evolution, Vol. 9, N.º 10, 01.10.2018, p. 2145-2154.

Resultado de la investigación: Contribución a RevistaArtículo

TY - JOUR

T1 - Graphs in phylogenetic comparative analysis

T2 - Anscombe's quartet revisited

AU - Revell, Liam J.

AU - Schliep, Klaus

AU - Valderrama, Eugenio

AU - Richardson, James E.

PY - 2018/10/1

Y1 - 2018/10/1

N2 - In 1973, the statistician Francis Anscombe used a clever set of bivariate datasets (now known as Anscombe's quartet) to illustrate the importance of graphing data as a component of statistical analyses. In his example, each of the four datasets yielded identical regression coefficients and model fits, and yet when visualized revealed strikingly different patterns of covariation between x and y. Phylogenetic comparative methods (the set of methodologies that use phylogenies, often combined with phenotypic trait data, to make inferences about evolution) are statistical methods too; yet visualizing the data and phylogeny in a sensible way that would permit us to detect unexpected patterns or unanticipated deviations from model assumptions is not a routine component of phylogenetic comparative analyses. Here, we use a quartet of phylogenetic datasets to illustrate that the same estimated parameters and model fits can be obtained from data that were generated using markedly different procedures—including pure Brownian motion evolution and randomly selected data uncorrelated with the tree. Just as in the case of Anscombe's quartet, when graphed the differences between the four datasets are quickly revealed. The intent of this article is to help build the general case that phylogenetic comparative methods are statistical methods and consequently that graphing or visualization should invariably be included as an essential step in our standard data analytical pipelines. Phylogenies are complex data structures and thus visualizing data on trees in a meaningful and useful way is a challenging endeavour. We recommend that the development of graphical methods for simultaneously visualizing data and tree should continue to be an important goal in phylogenetic comparative biology.

AB - In 1973, the statistician Francis Anscombe used a clever set of bivariate datasets (now known as Anscombe's quartet) to illustrate the importance of graphing data as a component of statistical analyses. In his example, each of the four datasets yielded identical regression coefficients and model fits, and yet when visualized revealed strikingly different patterns of covariation between x and y. Phylogenetic comparative methods (the set of methodologies that use phylogenies, often combined with phenotypic trait data, to make inferences about evolution) are statistical methods too; yet visualizing the data and phylogeny in a sensible way that would permit us to detect unexpected patterns or unanticipated deviations from model assumptions is not a routine component of phylogenetic comparative analyses. Here, we use a quartet of phylogenetic datasets to illustrate that the same estimated parameters and model fits can be obtained from data that were generated using markedly different procedures—including pure Brownian motion evolution and randomly selected data uncorrelated with the tree. Just as in the case of Anscombe's quartet, when graphed the differences between the four datasets are quickly revealed. The intent of this article is to help build the general case that phylogenetic comparative methods are statistical methods and consequently that graphing or visualization should invariably be included as an essential step in our standard data analytical pipelines. Phylogenies are complex data structures and thus visualizing data on trees in a meaningful and useful way is a challenging endeavour. We recommend that the development of graphical methods for simultaneously visualizing data and tree should continue to be an important goal in phylogenetic comparative biology.

UR - http://www.scopus.com/inward/record.url?scp=85052841574&partnerID=8YFLogxK

UR - http://www.scopus.com/inward/citedby.url?scp=85052841574&partnerID=8YFLogxK

U2 - 10.1111/2041-210X.13067

DO - 10.1111/2041-210X.13067

M3 - Article

AN - SCOPUS:85052841574

VL - 9

SP - 2145

EP - 2154

JO - Methods in Ecology and Evolution

JF - Methods in Ecology and Evolution

SN - 2041-210X

IS - 10

ER -