traits.faxes.cor {mFD}

Correlation between Traits and Axes

Description

Compute relationship between all traits and all axes of the functional space. For continuous trait a linear model is computed and r2 and p-value are returned. For other types of traits, a Kruskal-Wallis test is computed and eta2 statistics is returned. Option allows to plot trait-axis relationships with scatterplot and boxplot for continuous and non-continuous traits, respectively.

Usage

traits.faxes.cor(
  sp_tr,
  sp_faxes_coord,
  tr_nm = NULL,
  faxes_nm = NULL,
  plot = FALSE,
  name_file = NULL,
  color_signif = "darkblue",
  color_non_signif = "gray80",
  stop_if_NA = TRUE
)

Arguments

sp_tr

a data frame containing species as rows and traits as columns.

sp_faxes_coord

a matrix of species coordinates in a multidimensional functional space. Species coordinates have been retrieved thanks to tr.cont.fspace or quality.fspaces.

tr_nm

a vector gathering the names of traits (as in sp_tr) to consider. If NULL all traits are considered.

faxes_nm

a vector gathering the names of PCoA axes (as in sp_faxes_coord) to consider.

plot

a logical value indicating whether plot illustrating relations between trait and axes should be drawn. You can only plot relationships for up to 10 traits and/or 10 axes.

name_file

the file name (without extension) to save the plot as a 300 dpi JPEG file. Default is NULL which means plot is only displayed. If plot = FALSE this argument is ignored.

color_signif

an R color name or an hexadecimal code referring to the color of points when relationships between the trait and the axis is significant. Default is darkblue.

color_non_signif

an R color name or an hexadecimal code referring to the color of points when relationships between the trait and the axis are not significant. Default is gray80.

stop_if_NA

a logical value to stop or not the process if the sp_tr data frame contains NA. Functional measures are sensitive to missing traits. For further explanations, see the Note section. Default is TRUE.

Value

1 data frame with for each combination of trait and axis (rows), the name of the test performed, and the corresponding statistics and p-value. If plot = TRUE a multi-panel figure with traits as columns and axes as rows is also plotted. When relationships between trait and axis is significant the points are colored, else they remain grayish.

Author(s)

Nicolas Loiseau and Sebastien Villeger

Examples

# Load Species x Traits Data
data("fruits_traits", package = "mFD")

# Load Traits categories dataframe
data("fruits_traits_cat", package = "mFD")

# Compute Functional Distance
sp_dist_fruits <- mFD::funct.dist(sp_tr  = fruits_traits,
 tr_cat = fruits_traits_cat,
 metric = "gower",
 scale_euclid  = "scale_center",
 ordinal_var = "classic",
 weight_type = "equal",
 stop_if_NA  = TRUE)
  
# Compute Functional Spaces Quality (to retrieve species coordinates)
fspaces_quality_fruits <- mFD::quality.fspaces(
  sp_dist             = sp_dist_fruits, 
  maxdim_pcoa         = 10,
  deviation_weighting = "absolute",
  fdist_scaling       = FALSE,
  fdendro             = "average")
  
# Retrieve Species Coordinates
sp_faxes_coord_fruits <- fspaces_quality_fruits$details_fspaces$sp_pc_coord

# Compute Correlation between Traits and Functional Axes
mFD::traits.faxes.cor(
  sp_tr          = fruits_traits, 
  sp_faxes_coord = sp_faxes_coord_fruits, 
  tr_nm          = NULL, 
  faxes_nm       = NULL,
  name_file      = NULL, 
  color_signif   = "darkblue",
  color_non_signif = "gray80")