Parameter heatmap component of WFPHM

Usage

wfphm_hmpar_UI(id, tr_choices)

wfphm_hmpar_server(
  id,
  dataset,
  cat_var,
  par_var,
  visit_var,
  subjid_var,
  value_vars,
  sorted_x,
  tr_mapper,
  margin,
  show_x_ticks
)

Arguments

id

Shiny ID [character(1)]

dataset

[shiny::reactive(data.frame) | shinymeta::metaReactive(data.frame)]

It expects the following format:

it contains, at least, the columns specified in the parameters: cat_var, par_var, value_vars, visit_var and subjid_var
cat_var, par_var, visit_var and subjid_var columns are factors
value_vars must be numeric

cat_var, par_var, visit_var, subjid_var

[character(1)]

value_vars

[character(1+)]

possible colum values. If column is labelled, label will be displayed in the value menu

tr_mapper

[function(1+)]

named list containing a set of unary functions to transform the data. The name is the string shown in a selector and the value is the function to be applied according to details section.

margin

[numeric(4) | shiny::reactive(numeric(4)) | shinymeta::metaReactive(numeric(4))]

margin to be used on each of the sides. It must contain four entries named top, bottom, left and right

show_x_ticks

[logical(1)]

show x ticks in the parameter heatmap

Value

UI

A heatmap plot

Server

A list with one entry:

margin: similar to the margin parameter with the minimum margins for the current plot

Details

Data subsetting:

Allows selecting several values from the cat_var column, from now on cat_selection and several values from the par_var column from the subset of rows where par_cat is equal to cat_selection from now on par_selection.
- Menu labelled: Category and Parameter
Allows selecting between the columns defined in value_vars from now on value_selection.
- Menu labelled: Value
Allows selecting a value from visit_var column, from now on visit_selection.
- Menu labelled: Visit
Subsets the dataset rows where:
- visit_var equal to visit_selection
- par_var equal to par_selection
Subsets the dataset to par_var, subj_var and the value_selection.
If more than one row have the same combination subj_var and par_var an informative error indicating the plot cannot be created is shown.

Then the dataset is prepared to be passed to heatmap_D3:

subj_var becomes x column
the label attribute of x column is Subject ID
par_var becomes y column
value_selection becomes z column
z becomes label column

Completing the dataset:

Subset dataset will be completed in the following way. All non-present combination of the original levels of x and 'y' is are added with rows where:
x and y are equal to the missing combination
z is NA

Data outliers:

Allows setting two limits upper and lower for y value, values above or below in the subsetted dataset will be considered outliers. Rows considered outliers will have the column:
- label replaced by "x"
- color equal to "fill: white; font-weight: bold;"
Rows not considered outliers will have the column:
- color equal to NA

Data transformation:

Allows selecting between a set of functions as defined in tr_mapper from now on transformation_function.
- Menu labelled: Transformation
This transformation is applied to the subset dataset grouped by the y column. i.e. each row of the heatmap is transformed independently.
The function must be unary, and the unique argument will be the numerical values of a given row of the hetamap.

X axis ordering

x levels are ordered according to sorted_x

Then a call to heatmap_D3 is done with the following arguments:

data = subset dataset (as described above)
x_axis = S
y_axis = W
z_axis = E
margin is the parameter passed to this same function
palette is hardcoded to RColorBrewer::brewer.pal(n = 8, name = "Set2") after 8 categories colors are repeated
msg_func = NULL
quiet = TRUE

Functions

wfphm_hmpar_UI(): UI
wfphm_hmpar_server(): server