Skip to contents

Removes annotations where the names column is greater than an upper limit or less than a lower limit.

Usage

filter_range(
  column_name,
  upper_limit = Inf,
  lower_limit = -Inf,
  equal_to = TRUE,
  ...
)

Arguments

column_name

(character) The column name to filter.

upper_limit

(numeric, integer, function) The upper limit used for filtering. Can be a value, or a function that computes a value (e.g. mean). The default is Inf.

lower_limit

(numeric, integer, function) The lower limit used for filtering. Can be a value, or a function that computes a value (e.g. mean). The default is -Inf.

equal_to

(logical) Equal to limits. Allowed values are limited to the following:

  • "TRUE": Greater/less than or equal to the limits are excluded.

  • "FALSE": Greater/less than the limits are excluded.

The default is TRUE.

...

Additional slots and values passed to struct_class.

Value

A filter_range object with the following output slots:

filtered(annotation_source) Annotation_source after filtering.
flags(data.frame) A list of flags indicating which annotations were removed.

Inheritance

A filter_range object inherits the following struct classes:

[filter_range] -> [model] -> [struct_class]

Examples

M <- filter_range(
        column_name = "V1",
        upper_limit = Inf,
        lower_limit = -Inf,
        equal_to = FALSE)