edma_gdm: Growth difference
In psolymos/EDMAinR: Euclidean Distance Matrix Analysis in R

edma_gdm

R Documentation

Growth difference

Description

Growth matrix and growth difference matrix based inference based on Lele and Richtsmeier (1992, 1995).

Usage

edma_gm(a1, a2, ...)
get_gm(object, ...)
## S3 method for class 'edma_gm'
get_gm(object, sort=FALSE, level=0.95,
    what="all", ...)

edma_gdm(a1, a2, b1, b2, ...)
get_gdm(object, ...)
## S3 method for class 'edma_gdm'
get_gdm(object, sort=FALSE, level=0.95,
    what="all", ...)

## S3 method for class 'edma_gm'
print(x, ...)
## S3 method for class 'edma_gdm'
print(x, ...)
## S3 method for class 'edma_gm'
global_test(object, ...)
## S3 method for class 'edma_gdm'
global_test(object, ...)
## S3 method for class 'edma_gdm'
landmarks(x, ...)
## S3 method for class 'edma_gdm'
dimensions(x, ...)

## S3 method for class 'edma_gdm'
plot_ord(x, ...)
## S3 method for class 'edma_gdm'
plot_clust(x, ...)

Arguments

`a1, a2, b1, b2`	EDMA fit object to compare growths.
`x, object`	an EDMA GM or GDM objects.
`sort`	logical, if stacked distances are to be sorted, see Examples.
`level`	numeric, between 0 and 1, alpha level for confidence interval.
`what`	what part of the ford differences to return: `"all"`, `"less"` or `"greater"` than 1, `"signif"` or `"nonsignif"`.
`...`	other arguments passed to `edma_fdm`, like `ref_denom`.

Details

Growth matrix (GM) is calculated as the ratio of form matrices (FM) from the numerator and denominator objects following Lele and Richtsmeier (1992, 1995): GM(A1,A2) = FM(A2)/FM(A1). Form matrices are formed as pairwise Euclidean distances between landmarks from EDMA fit objects using the estimated mean forms.

Growth difference matrix (GDM) is calculated as GDM(A1,A2,B1,B2) = GM(B1,B2) / GM(A1,A2).

Inference and visualization is similar to how it is done for FDMs.

Value

edma_gm compares two EDMA fit objects and calculates GM.

edma_gdm compares 4 EDMA fit objects and calculates GDM.

The plot_ord and plot_clust produce plots based on dissimilarities among specimens in the 2 or 4 objects (for GM and GDM, respectively).

Author(s)

Peter Solymos, Subhash R. Lele, Theodore M. Cole, Joan T. Richtsmeier

References

Lele, S. R., and Richtsmeier, J. T., 1992. On comparing biological shapes: detection of influential landmarks. American Journal of Physical Anthropology 87:49–65. <doi:10.1002/ajpa.1330870106>

Lele, S. R., and Richtsmeier, J. T., 1995. Euclidean distance matrix analysis: confidence intervals for form and growth differences. American Journal of Physical Anthropology 98:73–86. <doi:10.1002/ajpa.1330980107>

Examples

file_a1 <- system.file("extdata/growth/CZEM_wt_global.xyz",
    package="EDMAinR")
file_a2 <- system.file("extdata/growth/CZP0_wt_global.xyz",
    package="EDMAinR")

l <- c("amsph", "bas", "loci", "lpto", "lsqu",
        "lsyn", "roci", "rpto", "rsqu", "rsyn")

a1 <- read_xyz(file_a1)[l,,]
a2 <- read_xyz(file_a2)[l,,]
a1
a2

fit_a1 <- edma_fit(a1, B=10)
fit_a2 <- edma_fit(a2, B=10)

## --- growth matrix ---

gm <- edma_gm(a1=fit_a1, a2=fit_a2, B=10)
gm
global_test(gm)
head(confint(gm))
head(get_gm(gm))
head(get_gm(gm, sort=TRUE, decreasing=TRUE))
head(get_gm(gm, sort=TRUE, decreasing=FALSE))

plot_ord(gm)
plot_clust(gm)
plot_test(gm)
plot_ci(gm)
plot_2d(gm)
if (interactive())
    plot_3d(gm)

## --- growth difference matrix ---

file_b1 <- system.file("extdata/growth/CZEM_mut_global.xyz",
    package="EDMAinR")
file_b2 <- system.file("extdata/growth/CZP0_mut_global.xyz",
    package="EDMAinR")

b1 <- read_xyz(file_b1)[l,,]
b2 <- read_xyz(file_b2)[l,,]
b1
b2

fit_b1 <- edma_fit(b1, B=10)
fit_b2 <- edma_fit(b2, B=10)

gdm <- edma_gdm(a1=fit_a1, a2=fit_a2, b1=fit_b1, b2=fit_b2, B=10)
gdm
global_test(gdm)
head(confint(gdm))
head(get_gdm(gdm))
head(get_gdm(gdm, sort=TRUE, decreasing=TRUE))
head(get_gdm(gdm, sort=TRUE, decreasing=FALSE))

plot_ord(gdm)
plot_clust(gdm)
plot_test(gdm)
plot_ci(gdm)
plot_2d(gdm) # need real data
if (interactive())
    plot_3d(gdm)

psolymos/EDMAinR documentation built on Aug. 25, 2023, 7:54 p.m.