Mark outlier rows.

mark_outlier_rows drops outlier rows.

mark_outlier_rows(
  population,
  variables,
  sample,
  method = "svd+iqr",
  outlier_col = "is_outlier",
  ...
)

Arguments

population	tbl with grouping (metadata) and observation variables.
variables	character vector specifying observation variables.
sample	tbl containing sample that is used by outlier removal methods to estimate parameters. sample has same structure as population. Typically, sample corresponds to controls in the experiment.
method	optional character string specifying method for outlier removal. There is currently only one option ("svd_iqr").
outlier_col	optional character string specifying the name for the column that will indicate outliers (in the output). Default "is_outlier".
...	arguments passed to outlier removal method.

Value

population with an extra column is_outlier.

Examples

suppressMessages(suppressWarnings(library(magrittr)))
population <- tibble::tibble(
  Metadata_group = sample(c("a", "b"), 100, replace = TRUE),
  Metadata_type = sample(c("control", "trt"), 100, replace = TRUE),
  AreaShape_Area = c(rnorm(98), 20, 30),
  AreaShape_Eccentricity = rnorm(100)
)
variables <- c("AreaShape_Area", "AreaShape_Eccentricity")
sample <- population %>% dplyr::filter(Metadata_type == "control")
population_marked <-
  cytominer::mark_outlier_rows(
    population,
    variables,
    sample,
    method = "svd+iqr"
  )
population_marked %>%
  dplyr::group_by(is_outlier) %>%
  dplyr::sample_n(3)
#> # A tibble: 6 x 5
#> # Groups:   is_outlier [2]
#>   Metadata_group Metadata_type AreaShape_Area AreaShape_Eccentricity is_outlier
#>   <chr>          <chr>                  <dbl>                  <dbl> <lgl>     
#> 1 a              trt                   0.390                -0.00636 FALSE     
#> 2 b              control               0.0767               -0.0596  FALSE     
#> 3 a              trt                   1.52                  0.695   FALSE     
#> 4 a              control              -2.33                 -1.91    TRUE      
#> 5 b              control              -1.56                  1.58    TRUE      
#> 6 a              trt                   2.30                  1.31    TRUE