Skip to contents

Discriminant Analysis of Principal Components for gen_tibble

gt_dapc.Rd

This function implements the Discriminant Analysis of Principal Components (DAPC, Jombart et al. 2010). This method describes the diversity between pre-defined groups. When groups are unknown, use gt_cluster_pca() to infer genetic clusters. See 'details' section for a succinct description of the method, and the vignette in the package adegenet ("adegenet-dapc") for a tutorial. This function returns objects of class adegenet::dapc which are compatible with methods from adegenet; graphical methods for DAPC are documented in adegenet::scatter.dapc (see ?scatter.dapc).

Usage

gt_dapc(
  x,
  pop = NULL,
  n_pca = NULL,
  n_da = NULL,
  loadings_by_locus = TRUE,
  pca_info = FALSE
)

Arguments

x

an object of class gt_pca, or its subclass gt_cluster_pca

pop

either a factor indicating the group membership of individuals; or an integer defining the desired k if x is a gt_cluster_pca; or NULL, if 'x' is a gt_cluster_pca and contain an element 'best_k', usually generated with gt_cluster_pca_best_k(), which will be used to select the clustering level.

n_pca

number of principal components to be used in the Discriminant Analysis. If NULL, k-1 will be used.

n_da

an integer indicating the number of axes retained in the Discriminant Analysis step.

loadings_by_locus

a logical indicating whether the loadings and contribution of each locus should be stored (TRUE, default) or not (FALSE). Such output can be useful, but can also create large matrices when there are a lot of loci and many dimensions.

pca_info

a logical indicating whether information about the prior PCA should be stored (TRUE, default) or not (FALSE). This information is required to predict group membership of new individuals using predict, but makes the object slightly bigger.

Value

an object of class adegenet::dapc

Details

The Discriminant Analysis of Principal Components (DAPC) is designed to investigate the genetic structure of biological populations. This multivariate method consists in a two-steps procedure. First, genetic data are transformed (centred, possibly scaled) and submitted to a Principal Component Analysis (PCA). Second, principal components of PCA are submitted to a Linear Discriminant Analysis (LDA). A trivial matrix operation allows to express discriminant functions as linear combination of alleles, therefore allowing one to compute allele contributions. More details about the computation of DAPC are to be found in the indicated reference.

References

Jombart T, Devillard S and Balloux F (2010) Discriminant analysis of principal components: a new method for the analysis of genetically structured populations. BMC Genetics 11:94. doi:10.1186/1471-2156-11-94 Thia, J. A. (2023). Guidelines for standardizing the application of discriminant analysis of principal components to genotype data. Molecular Ecology Resources, 23, 523–538. https://doi.org/10.1111/1755-0998.13706