Estimate the maximum number of groups in DAPC analysis

Usage

find_max_nclust(
  x,
  threshold,
  max.nclust,
  nperm = 100,
  method = "kmeans",
  stat = "BIC",
  criterion = "diffNgroup",
  subset = 100,
  confidence.level = c(0.7, 0.8, 0.9, 0.95, 0.99)
)

Arguments

x: A data.frame or matrix object containing eigenvectors by sites.
threshold: Scalar. The number of eigenvectors used to perform classification.
max.nclust: Scalar indicating the maximum number of groups to be evaluated.
nperm: Scalar. Number of times classification will be performed.
method: c("kmeans","ward"). See find.clusters of adegenet package.
stat: c("BIC", "AIC", or "WSS"). See find.clusters of adegenet package.
criterion: c("diffNgroup", "min","goesup", "smoothNgoesup", or "goodfit"). See find.clusters of adegenet package.
subset: number of cells used in the analysis. It is particularly important whenever the total number of cells is large (> 1000).
confidence.level: threshold confidence level used to estimate congruence in the classification pattern.

Value

Matrix containing congruence values ranging between 0-1 for each max.nclust value (see Arguments) and confidence level.

Examples

if (FALSE) {
data(regions)
evovectors <- regions$PCPS$vectors # eigenvectors by site
find_max_nclust(x=evovectors, threshold=3,max.nclust=10)
}