TESS function to do everything

Usage

tess_do_everything(
  gen,
  coords,
  grid = NULL,
  Kvals = 1:10,
  K_selection = "manual",
  plot_method = "maxQ",
  col_breaks = 20,
  minQ = 0.1,
  tess_method = "projected.ls",
  lambda = 1,
  ploidy = 2,
  correct_kriged_Q = TRUE,
  quiet = FALSE,
  model = c("Sph", "Exp", "Gau", "Ste")
)

Arguments

gen: genotype dosage matrix (rows = individuals & columns = snps) or vcfR object
coords: coordinates of samples as sf points, a two-column matrix, or a data.frame representing x and y coordinates (see Details for important information about projections)
grid: Optional SpatRaster for kriging (defaults to NULL)
Kvals: vector of K values to test
K_selection: how to perform K selection ("manual" to enter into console (default) or "auto" for automatic selection based on bestK)
plot_method: method for making rainbow map of kriged layers (options: "maxQ" to only plot the max Q value for each cell (default), "allQ" to plot all Q values greater than minQ, "maxQ_poly" or "allQ_poly" to create the plots as previously described, but as polygons for each K instead of continuous Q values)
col_breaks: number of breaks for plotting (defaults to 20)
minQ: threshold for minimum Q value for rainbow plotting if method = "all" is used (defaults to 0.10)
tess_method: the type of TESS method to be run ("projected.ls" for projected least squares algorithm (default) or "qp" for quadratic programming algorithm)
lambda: numeric value for the spatial regularization parameter. The default value lambda = 1 attributes equal weights to the loss function and to the penalty function.
ploidy: ploidy of data (defaults to 2)
correct_kriged_Q: whether to correct kriged Q values so values greater than 1 are set to 1 and values less than 0 are set to 0 (defaults to TRUE)
quiet: whether to operate quietly and suppress the output of tables and figures (defaults to FALSE)
model: variogram model(s) to test during automatic kriging. Passed to autoKrige. Defaults to c("Sph", "Exp", "Gau", "Ste"). To restrict fitting to specific models, provide one or more model names (e.g., c("Sph", "Exp")).

Value

list with all TESS results, final K value, and final kriged raster

Details

TESS is run using the tess3r package: Caye, K., François, O. (2016). tess3r: Inference of Spatial Population Genetic Structure. R package version 1.1.0. See also: Caye, K., Deist, T.M., Martins, H., Michel, O., François, O. (2016). TESS3: fast inference of spatial population structure and genome scans for selection. Mol. Ecol. Res. 16(2):540-548. https://doi.org/10.1111/1755-0998.12471

Coordinates and rasters should be in a projected (planar) coordinate system. Therefore, spherical systems (including latitute-longitude coordinate systems) should be projected prior to use. Transformation can be performed using st_set_crs for coordinates or project for rasters (see vignette for more details).