Generalized Procrustes Analysis (GPA) aligns our landmark
configurations to a common coordinate system, and allows us to extract
shape (and size) data from them. The function gpagen
performs this procedure. Here we demonstrate its use for landmark
data:
library(geomorph)
## Loading required package: RRPP
## Loading required package: rgl
## Loading required package: Matrix
data(plethodon)
pleth.gpa <- gpagen(plethodon$land, print.progress = F)
summary(pleth.gpa)
##
## Call:
## gpagen(A = plethodon$land, print.progress = F)
##
##
##
## Generalized Procrustes Analysis
## with Partial Procrustes Superimposition
##
## 12 fixed landmarks
## 0 semilandmarks (sliders)
## 2-dimensional landmarks
## 2 GPA iterations to converge
##
##
## Consensus (mean) Configuration
##
## X Y
## 1 0.15233235 -0.025236647
## 2 0.19328979 -0.095041344
## 3 -0.03370053 -0.006929874
## 4 -0.28182427 -0.089370882
## 5 -0.31072667 -0.057833070
## 6 -0.32600020 -0.032082160
## 7 -0.31757270 0.040056685
## 8 -0.18824427 0.100347118
## 9 0.02159274 0.098853343
## 10 0.18946790 0.074940119
## 11 0.35213101 0.061515271
## 12 0.54925485 -0.069218559
Using summary
provides us with a general description of
the analysis we performed. Additionally, there are quite a few
components to our GPA object: the most important of which are
$coords
and $Csize
$
attributes(pleth.gpa)
## $names
## [1] "coords" "Csize" "iter" "rot.pts" "consensus"
## [6] "procD" "p" "k" "nsliders" "nsurf"
## [11] "points.VCV" "points.var" "data" "Q" "slide.method"
## [16] "call"
##
## $class
## [1] "gpagen"
plot(pleth.gpa)
plotAllSpecimens(pleth.gpa$coords, links = plethodon$links)
Use of plot.param
can facilitate additional flexibility
(change point or link colors, thickness, etc.)