Input general data in a teal application

NEST CoreDev

2022-04-20

Introduction

Teal applications are not restricted to CDISC-standard data. Although many teal modules included with NEST are designed for CDISC data, those in the library teal.modules.general have been designed to work with non-relational data.

For example this application uses the standard iris and mtcars datasets:

library(teal)

app <- init(
  data = teal_data(
    dataset("IRIS", iris, code = "IRIS <- iris"),
    dataset("CARS", mtcars, code = "CARS <- mtcars")
  ),
  modules = example_module()
)

if (interactive()) {
  shinyApp(app$ui, app$server)
}

For more information, see documentation in teal.data.

Delayed Data Loading (DDL)

To learn more about DDL, visit vignette("delayed-data-loading", package = "teal.data"). All the features of DDL are available for general datasets:

library(teal)

person_generator <- function() {
  return(
    data.frame(
      ID = factor(1:8),
      AGE = c(40, 23, 56, 11, 17, 71, 23, 56)
    )
  )
}

pet_generator <- function() {
  return(
    data.frame(
      ID = factor(1:10),
      TYPE = rep(c("CAT", "DOG"), 5),
      COLOR = c("GINGER", rep("BROWN", 5), rep("BLACK", 4)),
      PERSON_ID = factor(c(5, 4, 3, 3, 3, 1, 8, 1, 2, 2))
    )
  )
}


app <- init(
  data = teal_data(
    fun_dataset_connector("PERSON", fun = person_generator, keys = "ID") %>%
      mutate_dataset("PERSON$SEX <- rep(c('M', 'F'), 4)"),
    fun_dataset_connector("PETS", fun = pet_generator, keys = "ID")
  ) %>%
    mutate_join_keys("PERSON", "PETS", c("ID" = "PERSON_ID")),
  modules = example_module()
)

if (interactive()) {
  shinyApp(app$ui, app$server)
}

For more information about preprocessing, reproducibility, relationships between datasets and DDL, please refer to the teal.data package.