Permutation tests for phylogenetic comparative analyses of highdimensional shape data: What you shuffle matters
Evaluating statistical trends in high‐dimensional phenotypes poses challenges for comparative biologists, because the high‐dimensionality of the trait data relative to the number of species can prohibit parametric tests from being computed. Recently, two comparative methods were proposed to circumve... Full description
Journal Title:  Evolution 201503, Vol.69 (3), p.823829 
Main Author:  Adams, Dean C 
Other Authors:  Collyer, Michael L 
Format:  Electronic Article 
Language: 
English 
Subjects:  
Publisher:  United States: Blackwell Publishing Ltd 
ID:  ISSN: 00143820 
Link:  https://www.ncbi.nlm.nih.gov/pubmed/25641367 
Zum Text: 
SendSend as email
Add to Book BagAdd to Book Bag
Staff View
recordid:  cdi_proquest_miscellaneous_1664447653 
title:  Permutation tests for phylogenetic comparative analyses of highdimensional shape data: What you shuffle matters 
format:  Article 
creator: 

subjects: 

ispartof:  Evolution, 201503, Vol.69 (3), p.823829 
description:  Evaluating statistical trends in high‐dimensional phenotypes poses challenges for comparative biologists, because the high‐dimensionality of the trait data relative to the number of species can prohibit parametric tests from being computed. Recently, two comparative methods were proposed to circumvent this difficulty. One obtains phylogenetic independent contrasts for all variables, and statistically evaluates the linear model by permuting the phylogenetically independent contrasts (PICs) of the response data. The other uses a distance‐based approach to obtain coefficients for generalized least squares models (D‐PGLS), and subsequently permutes the original data to evaluate the model effects. Here, we show that permuting PICs is not equivalent to permuting the data prior to the analyses as in D‐PGLS. We further explain why PICs are not the correct exchangeable units under the null hypothesis, and demonstrate that this misspecification of permutable units leads to inflated type I error rates of statistical tests. We then show that simply shuffling the original data and recalculating the independent contrasts with each iteration yields significance levels that correspond to those found using D‐PGLS. Thus, while summary statistics from methods based on PICs and PGLS are the same, permuting PICs can lead to strikingly different inferential outcomes with respect to statistical and biological inferences. 
language:  eng 
source:  
identifier:  ISSN: 00143820 
fulltext:  no_fulltext 
issn: 

url:  Link 
@attributes 
 
PrimoNMBib 
