The goal of GuiRBioDiv is to provide some wrapper functions to speed-up alpha diversity analyses of metabarcoding datasets.

It features:

• alpha_residual for correction of sequencing depth bias for alpha diversity indices with linear models
• endemic to identify species/OTUs/ASVs uniquely found in one geographic region
• transformTukey2 to transform numeric vectors to more normally distributed values with Tukey’s ladder of power method

and few more …

GuiRBioDiv can be installed from GitHub like so:

install.packages("remotes")
remotes::install_github("lentendu/GuiRBioDiv")

Compute Hill numbers and correct for sequencing depth bias:

library(GuiRBioDiv)
alphadiv<-alpha_residual(mat_exa, index="hill", inform=T)
#> [1] "index hill0 best linear correlation to counts is with no transformation: Adj. R2 = 0.49 ;  p.value = 1.46e-02"
#> [1] "index hill1 is not linearly correlated to counts, neither to log or sqrt transformed counts"
#> [1] "index hill2 is not linearly correlated to counts, neither to log or sqrt transformed counts"