Tutorial R : Spatial Analysis

Tutorial using the free softwarePerforming a spatial Principal ComponentAnalysis using the R softwareT. JOMBARTNovember 2006For any questions or comments, please send an email to:jombart@biomserv.univ-lyon1.fr.Contents1 Introduction 22 Building a connection network 43 Performing the analyses 53.1 Testing for the existence of spatial structures . . . . . . . . . . . . . 63.2 Identifying and displaying the spatial structures . . . . . . . . . . . 83.2.1 The Principal Component Analysis . . . . . . . . . . . . . . 83.2.2 The spatial Principal Component Analysis . . . . . . . . . . 101T. JOMBART, A sPCA tutorial using R1 IntroductionThistutorialshowshowtoperformaspatial Principal Component Analysis (sPCA,Jombart et al., submitted), starting from a data set of allelic frequencies of geo-referenced individuals or groups of individuals. The second illustration presentedinthepreviouslycitedarticleisreproducedbelow. Itinvolvesgroupsofindividuals(cat colonies) data, but the process is exactly the same using individuals data.Before starting, please download the files Nancy cats.rda and map.ppm athttp://pbil.univ-lyon1.fr/R/html/additifs.html. The first file is a dataset with R format (’rda’), and the second one is a picture of the sampled area.We assume that both files were saved in your R working directory (which can beknown and set using the commands getwd() and setwd()). Several R packagesare also used to perform the analyses. These can be installed using ...