Skip to contents

Process vertebral measurements and PC scores of 2D or 3D geometric morphometric data

Source: R/process_input.R
process_gmPC.Rd

process_gmPC() initializes the analysis workflow by processing vertebral measurements and already computed PC scores from 2D or 3D geometric morphometric datasets into an object usable by MorphoRegions. Such processing includes identifying the vertebra indices, measurements, and PC scores.

Usage

process_gmPC(data, pos = "names", pcscores, eigenvals, specimens = NULL)

Arguments

data

a 3D array (\(p \times k \times n\)) of Procrustes aligned landmarks where \(p\) is the number of of landmark points per vertebra, \(k\) is the number of landmark dimensions (2D or 3D), and \(n\) is the number of vertebrae sampled for the specimen.

pos

either character string "names" in which case the function will extract vertebral position info form the names of the array (using dimnames(data)[[3]]), or a vector containing numeric values of vertebral position. Default is "names".

pcscores

a matrix or data frame containing PC scores of each vertebra.

eigenvals

a numeric vector containing the eigenvalues of each PC axis.

specimens

if data contains multiple specimens, a vector of specimen names of length equal to the number of vertebrae in data.

Value

A regions_pco object, which contains user-provided eigenvectors in the scores component and eigenvalues in the eigen.val component. The original dataset, including landmark coordinates and positional information, is stored in the data attribute.

Details

Unlike process_measurements(), process_gmPC() does not fill in missing values and these should have been removed or replaced by numeric values before running process_gmPC().

See also

  • process_PC() for processing PC scores from traditional morphometric datasets

  • svdPCO() for computing principal coordinate axes from processed vertebra data.

  • plot.regions_pco() for plotting PCO axes

Examples

# Load coordinates of aligned landmarks, PC scores, and
# eigenvalues; vertebra index as names of array elements
# of aligned landmarks
data("seal_gmm")

# Process PC scores and morphological data:
seal_pca <- process_gmPC(data = seal_gmm$coord_gpa,
                         pcscores = seal_gmm$scores,
                         eigenvals = seal_gmm$eigenvals)

# Process multiple datasets; specimen names in 'specimens'
# vector, vertebrae index in 'vertebrae' vector
data("wolves_gmm")

# Process PC scores and geometric morphometric dataset:
wolves_pca <- process_gmPC(data = wolves_gmm$coord_gpa,
                           pos = wolves_gmm$vertebrae,
                           pcscores = wolves_gmm$scores,
                           eigenvals = wolves_gmm$eigenvals,
                           specimens = wolves_gmm$specimens)
#> Warning: `pos` supplied as numeric vector; assuming its order matches the order of
#> `data`.
#> Warning: The row names of `pcscores` do not match `pos`; assuming row order matches
#> `pos`.