Introduction
teal
extends the shiny
framework, enabling
the creation of interactive GUI applications using the R
.
shiny
, and teal
facilitate the development of
extensive applications through combining small, decoupled modules. The
teal.modules.general
package consist of collection of
modules essential for developing teal
applications. It is
“general” in the sense that the intended functions of these modules are
more fundamental. This contrasts with the more specialized focus on
clinical data found in the teal.modules.clinical
package.
The modules from teal.modules.general
can be used in
conjunction with modules from teal.modules.clinical
and /
or other shiny
modules to build a large teal
/
shiny
app.
The concepts presented here require knowledge about the core features
of teal
, specifically on how to launch a teal
application and how to pass data into it. Therefore, it is highly
recommended to refer to the README
file and the introductory vignette
of the teal
package.
See also teal.modules.clinical
’s README
.
Main features
There are five areas of data science that
teal.modules.general
provides tools and solutions (modules)
for:
- viewing data in tabular formats
- visualizing data in plots and graphs
- viewing data and files in a directory
- examining missing and extreme values in data
- performing data analysis
See package
functions / modules.
Example application
A simple application featuring the tm_variable_browser()
module:
# load libraries
library(teal.modules.general)
library(teal.widgets)
library(sparkline)
# teal_data object
data <- teal_data()
data <- within(data, {
ADSL <- rADSL
ADTTE <- rADTTE
})
join_keys(data) <- default_cdisc_join_keys[names(data)]
# tm_variable_browser module
tm_variable_browser_module <- tm_variable_browser(
label = "Variable browser",
ggplot2_args = ggplot2_args(
labs = list(subtitle = "Plot generated by Variable Browser Module")
)
)
# initialize the app
app <- init(
data = data,
modules = modules(tm_variable_browser_module)
)
shinyApp(app$ui, app$server)
Try it out in Shinylive
Open
in Shinylive
Let’s break the above app into pieces:
1: Load the necessary libraries and data.
library(teal.modules.general)
library(teal.widgets)
2: Construct a teal_data
object containing that will
serve as the source of data for the teal
app.
teal_data
not only encapsulates the data for the app, but
it also houses the code required to create the data to maintain
reproducibility.
To do this, we create an empty teal_data
object and
evaluate code to produce the data within the teal_data
object, so both the code and data are stored together.
Following this, we set the datanames
and
join_keys
.
data <- teal_data()
data <- within(data, {
ADSL <- rADSL
ADTTE <- rADTTE
})
join_keys(data) <- default_cdisc_join_keys[names(data)]
3: Initialize a teal
application with specified data and
modules, in this case, the module: tm_variable_browser
,
datasets:ADSL
and ADTTE
.
shiny::shinyApp()
use the ui
and
server
component to initialize the teal
app.
tm_variable_browser_module <- tm_variable_browser(
# module name to display in the GUI
label = "Variable browser",
# this argument takes a set of arguments to pass to ggplot2.
# the arguments must have the same names as its ggplot2 counterpart, e.g. `subtitle`
ggplot2_args = ggplot2_args(
labs = list(subtitle = "Plot generated by Variable Browser Module")
)
)
app <- init(
data = data,
modules = modules(tm_variable_browser_module)
)
shinyApp(app$ui, app$server)
In a teal
app, data and modules are decoupled. In the
app above:
The app developer specified the data and assigned it to the
data
argument.
The app developer specified the module and assigned it to the
modules
argument.
The init
function took these arguments and returned
a list containing ui
and server
object, which
can be demonstrated by running:
## [1] "list"
## [1] "ui" "server"