Permanova (adonis) on permutations of rarefaction even depth

Permanova are computed on a given number of rarefaction with different seed number. This reduce the risk of a random drawing of an exceptional situation of an unique rarefaction. Each permutation rarefies physeq to even depth (using MiscMetabar::rarefy_even_depth_pq()) and runs MiscMetabar::adonis_pq(); results are summarized across permutations.

Usage

adonis_rarperm_pq(
  physeq,
  formula,
  dist_method = "bray",
  merge_sample_by = NULL,
  na_remove = FALSE,
  rarefy_nb_seqs = FALSE,
  verbose = TRUE,
  nperm = 99,
  progress_bar = TRUE,
  quantile_prob = 0.975,
  sample.size = min(phyloseq::sample_sums(physeq)),
  ...
)

Arguments

physeq: (required) A phyloseq-class object obtained using the phyloseq package.
formula: (required) the right part of a formula for vegan::adonis2(). Variables must be present in the physeq@sam_data slot.
dist_method: (default "bray") the distance used. See phyloseq::distance() for all available distances or run phyloseq::distanceMethodList(). For aitchison and robust.aitchison distance, vegan::vegdist() function is directly used.
merge_sample_by: a vector to determine which samples to merge using the MiscMetabar::merge_samples2() function. Need to be in physeq@sam_data.
na_remove: (logical, default FALSE) If set to TRUE, remove samples with NA in the variables set in formula.
rarefy_nb_seqs: (logical, default FALSE) Rarefy each sample (before merging if merge_sample_by is set) using MiscMetabar::rarefy_even_depth_pq() inside MiscMetabar::adonis_pq().
verbose: (logical, default TRUE) If TRUE, prompt some messages.
nperm: (int, default 99) The number of permutations to perform.
progress_bar: (logical, default TRUE) Do we print progress during the calculation?
quantile_prob: (float, [0:1]) the value to compute the quantile. Minimum quantile is computed using 1 - quantile_prob.
sample.size: (int) A single integer value equal to the number of reads being simulated, also known as the depth. See phyloseq::rarefy_even_depth() and MiscMetabar::rarefy_even_depth_pq().
...: Other params to be passed on to MiscMetabar::adonis_pq() function.

Value

A list of three dataframe representing the mean, the minimum quantile and the maximum quantile value for adonis results. See MiscMetabar::adonis_pq().

Author

Adrien Taudière

Examples

# \donttest{
if (requireNamespace("vegan")) {
  data_fungi_woNA <-
    subset_samples(data_fungi_mini, !is.na(Time) & !is.na(Height))
  adonis_rarperm_pq(data_fungi_woNA, "Time*Height", na_remove = TRUE, nperm = 3)
}
#> 
  |                                                        
  |                                                  |   0%
#> Taxa are now in columns.
#> Removing NA from Time
#> Removing NA from Height
#> 
  |                                                        
  |=================                                 |  33%
#> Taxa are now in columns.
#> Removing NA from Time
#> Removing NA from Height
#> 
  |                                                        
  |=================================                 |  67%
#> Taxa are now in columns.
#> Removing NA from Time
#> Removing NA from Height
#> 
  |                                                        
  |==================================================| 100%
#> $mean
#>          Df  SumOfSqs         R2         F Pr(>F)
#> Model     5  2.346505 0.06649319 0.9844575  0.569
#> Residual 69 32.942384 0.93350681        NA     NA
#> Total    74 35.288889 1.00000000        NA     NA
#> 
#> $quantile_min
#>          Df  SumOfSqs         R2         F  Pr(>F)
#> Model     5  2.070369 0.05846765 0.8569665 0.11735
#> Residual 69 32.531964 0.92132923        NA      NA
#> Total    74 35.130000 1.00000000        NA      NA
#> 
#> $quantile_max
#>          Df  SumOfSqs         R2        F Pr(>F)
#> Model     5  2.778036 0.07867077 1.178625 0.8821
#> Residual 69 33.340964 0.94153235       NA     NA
#> Total    74 35.421333 1.00000000       NA     NA
#> 
# }