Post hoc pairwise comparisons for multiple groups test.

Source: R/post-hoc-test.R

Multiple group test, such as anova and Kruskal-Wallis rank sum test, can be used to uncover the significant feature among all groups. Post hoc tests are used to uncover specific mean differences between pair of groups.

run_posthoc_test(
  ps,
  group,
  transform = c("identity", "log10", "log10p"),
  norm = "TSS",
  norm_para = list(),
  conf_level = 0.95,
  method = c("tukey", "games_howell", "scheffe", "welch_uncorrected")
)

Arguments

ps

a phyloseq::phyloseq object

group

character, the variable to set the group

transform

character, the methods used to transform the microbial abundance. See transform_abundances() for more details. The options include:

  • "identity", return the original data without any transformation (default).

  • "log10", the transformation is log10(object), and if the data contains zeros the transformation is log10(1 + object).

  • "log10p", the transformation is log10(1 + object).

norm

the methods used to normalize the microbial abundance data. See normalize() for more details. Options include:

  • a integer, e.g. 1e6 (default), indicating pre-sample normalization of the sum of the values to 1e6.

  • "none": do not normalize.

  • "rarefy": random subsampling counts to the smallest library size in the data set.

  • "TSS": total sum scaling, also referred to as "relative abundance", the abundances were normalized by dividing the corresponding sample library size.

  • "TMM": trimmed mean of m-values. First, a sample is chosen as reference. The scaling factor is then derived using a weighted trimmed mean over the differences of the log-transformed gene-count fold-change between the sample and the reference.

  • "RLE", relative log expression, RLE uses a pseudo-reference calculated using the geometric mean of the gene-specific abundances over all samples. The scaling factors are then calculated as the median of the gene counts ratios between the samples and the reference.

  • "CSS": cumulative sum scaling, calculates scaling factors as the cumulative sum of gene abundances up to a data-derived threshold.

  • "CLR": centered log-ratio normalization.

norm_para

arguments passed to specific normalization methods

conf_level

confidence level, default 0.95

method

one of "tukey", "games_howell", "scheffe", "welch_uncorrected", defining the method for the pairwise comparisons. See details for more information.

Value

a postHocTest object

Examples

data(enterotypes_arumugam)
ps <- phyloseq::subset_samples(
    enterotypes_arumugam,
    Enterotype %in% c("Enterotype 3", "Enterotype 2", "Enterotype 1")
) %>%
    phyloseq::subset_taxa(Phylum == "Bacteroidetes")
pht <- run_posthoc_test(ps, group = "Enterotype")
pht
#> postHocTest-class object
#> Pairwise test result of 13  features,  DataFrameList object, each DataFrame has five variables:
#>         comparisons    : pair groups to test which separated by '-'
#>         diff_mean: difference in mean proportions
#>         pvalue        : post hoc test p values
#>         ci_lower : lower confidence interval
#>         ci_upper : upper confidence interval
#> Posthoc multiple comparisons of means  using  tukey  method