centerAtMax: centers profile of x and y. default is to center by region...

In jrboyd/seqsetvis: Set Based Visualizations for Next-Gen Sequencing Data

centers profile of x and y. default is to center by region but across all samples.

Description

centerAtMax locates the coordinate x of the maximum in y and shifts x such that it is zero at max y.

Usage

centerAtMax(
  dt,
  x_ = "x",
  y_ = "y",
  by_ = "id",
  view_size = NULL,
  trim_to_valid = TRUE,
  check_by_dupes = TRUE,
  x_precision = 3,
  replace_x = TRUE
)

Arguments

dt	data.table
x_	the variable name of the x-values. default is 'x'
y_	the variable name of the y-values default is 'y'
by_	optionally, any variables that provide grouping to the data. default is none. see details.
view_size	the size in x_ to consider for finding the max of y_. if length(view_size) == 1, range will be c(-view_size, view_size). if length(view_size) > 1, range will be range(view_size). default value of NULL uses complete range of x.
trim_to_valid	valid x_ values are those with a set y_ value in all by_ combinations
check_by_dupes	default assumption is that there should be on set of x_ for a by_ instance. if this is not the case and you want to disable warnings about set this to FALSE.
x_precision	numerical precision of x, default is 3.
replace_x	logical, default TRUE. if TRUE x_ will be replaced with position relative to summit. if FALSE x_ will be preserved and x_summitPosition added.

Details

character. by_ controls at the level of the data centering is applied. If by_ is "" or NULL, a single max position will be determined for the entire dataset. If by is "id" (the default) then each region will be centered individually across all samples.

Value

data.table with x (or xnew if replace_x is FALSE) shifted such that x = 0 matches the maximum y-value define by by_ grouping

Examples

data(CTCF_in_10a_profiles_gr)
centerAtMax(CTCF_in_10a_profiles_gr, y_ = 'y', by_ = 'id',
  check_by_dupes = FALSE)
#it's a bit clearer what's happening with trimming disabled
#but results are less useful for heatmaps etc.
centerAtMax(CTCF_in_10a_profiles_gr, y_ = 'y', by_ = 'id',
  check_by_dupes = FALSE, trim_to_valid = FALSE)
#specify view_size to limit range of x values considered, prevents
#excessive data trimming.
centerAtMax(CTCF_in_10a_profiles_gr, y_ = 'y', view_size = 100, by_ = 'id',
check_by_dupes = FALSE)

jrboyd/seqsetvis documentation built on Dec. 10, 2024, 11:23 a.m.