Mutual finding plot — mutual

Plots number of differentially abundant features mutually found by defined number of methods, colored by the differential abundance direction and separated by comparison.

Usage

mutual_plt(
  rec,
  count_cutoff = NULL,
  comparisons = NULL,
  steps = steps_ids(rec, type = "da"),
  top_n = 20,
  target = NULL,
  contrast_id = NULL
)

Arguments

rec: A prepped Recipe object.
count_cutoff: Indicates the minimum number of methods in which an OTU must be present (Default: NULL). If count_cutoff is NULL count_cutoff is equal to length(steps_ids(rec, "da")) * 2 / 3.
comparisons: By default, this function plots all comparisons. However, if the user indicates the comparison or comparisons of interest, only the selected ones will be plotted.
steps: Character vector with step_ids to take in account. Default all "da" methods.
top_n: Maximum number of taxa to represent. Default: 20.
target: Optional modeled target to include.
contrast_id: Optional modeled contrast identifier to include.

Value

ggplot2, or invisible NULL with an informational message when no significant taxa are available for the selected hypotheses.

Examples

data(test_prep_rec)

mutual_plt(test_prep_rec)
#> ℹ `count_cutoff` set to 1.
#> ! Taxa present in selected methods are greater than the cutoff `top_n` = 20.
#> ℹ The top 20 significant taxa with the greatest overlap between methods will be
#>   used.


## Require agreement among all methods
mutual_plt(
  test_prep_rec,
  count_cutoff = length(steps_ids(test_prep_rec, "da"))
)


## Show a single comparison
comparison <- tidy_results(test_prep_rec)$comparison[[1]]
mutual_plt(test_prep_rec, comparisons = comparison)
#> ℹ `count_cutoff` set to 1.
#> ! Taxa present in selected methods are greater than the cutoff `top_n` = 20.
#> ℹ The top 20 significant taxa with the greatest overlap between methods will be
#>   used.


## Exclude the first method from the plot
mutual_plt(test_prep_rec, steps = steps_ids(test_prep_rec, "da")[-1])
#> ℹ `count_cutoff` set to 1.
#> ! Taxa present in selected methods are greater than the cutoff `top_n` = 20.
#> ℹ The top 20 significant taxa with the greatest overlap between methods will be
#>   used.