createLandingPage: Create a landing page

View source: R/createLandingPage.R

createLandingPageR Documentation

Create a landing page

Description

Define a function to create a landing page in which users can specify or upload SummarizedExperiment objects.

Usage

createLandingPage(
  seUI = NULL,
  seLoad = NULL,
  initUI = NULL,
  initLoad = NULL,
  requireButton = TRUE
)

Arguments

seUI

Function that accepts a single id argument and returns a UI element for specifying the SummarizedExperiment.

seLoad

Function that accepts the input value of the UI element from seUI and returns a SummarizedExperiment object.

initUI

Function that accepts a single id argument and returns a UI element for specifying the initial state.

initLoad

Function that accepts the input value of the UI element from initUI and returns a list of Panels.

requireButton

Logical scalar indicating whether the app should require an explicit button press to initialize, or if it should initialize upon any modification to the UI element in seUI.

Details

By default, this function creates a landing page in which users can upload an RDS file containing a SummarizedExperiment, which is subsequently read by readRDS to launch an instance of iSEE. However, any source of SummarizedExperiment objects can be used; for example, we can retrieve them from databases by modifying seUI and seLoad appropriately.

The default landing page also allows users to upload a RDS file containing a list of Panels that specifies the initial state of the iSEE instance (to be used as the initial argument in iSEE). Again, any source can be used to create this list if initUI and initLoad are modified appropriately.

The UI elements for the SummarizedExperiment and the initial state are named "se" and "initial" respectively. This can be used in Shiny bookmarking to initialize an iSEE in a desired state by simply clicking a link, provided that requireButton=FALSE so that reactive expressions are immediately triggered upon setting se= and initial= in the URL. We do not use bookmarking to set all individual iSEE parameters as we will run afoul of URL character limits.

Value

A function that generates a landing page upon being passed to iSEE as the landingPage argument.

Defining a custom landing page

We note that createLandingPage is just a limited wrapper around the landing page API. In iSEE, landingPage can be any function that accepts the following arguments:

  • FUN, a function to initialize the iSEE observer architecture. This function expects to be passed:

    • SE, a SummarizedExperiment object.

    • INITIAL, a list of Panel objects describing the initial application state. If NULL, the initial state from initial in the top-level iSEE call is used instead.

    • TOUR, a data.frame containing a tour to be attached to the app - see defaultTour for an example. If NULL (the default), no tour is added.

    • COLORMAP, an ExperimentColorMap object that defines the colormaps to use in the application.

  • input, the Shiny input list.

  • output, the Shiny output list.

  • session, the Shiny session object.

The landingPage function should define a renderUI expression that is assigned to output$allPanels. This should define a UI that contains all widgets necessary for a user to set up an iSEE session interactively. We suggest that all UI elements have IDs prefixed with "initialize_INTERNAL" to avoid conflicts.

The function should also define observers to respond to user interactions with the UI elements. These are typically used to define a SummarizedExperiment object and an input state as a list of Panels; any one of these observers may then call FUN on those arguments to launch the main iSEE instance.

Note that, once the main app is launched, the UI elements constructed here are lost and observers will never be called again. There is no explicit “unload” mechanism to return to the landing page from the main app, though a browser refresh is usually sufficient.

Author(s)

Aaron Lun

Examples

createLandingPage()

# Alternative approach, to create a landing page
# that opens one of the datasets from the scRNAseq package.
library(scRNAseq)
all.data <- ls("package:scRNAseq")
all.data <- all.data[grep("Data$", all.data)]

lpfun <- createLandingPage(
    seUI=function(id) selectInput(id, "Dataset:", choices=all.data),
    seLoad=function(x) get(x, as.environment("package:scRNAseq"))()
)

app <- iSEE(landingPage=lpfun)
if (interactive()) {
  shiny::runApp(app, port=1234)
}


csoneson/SEE documentation built on Dec. 13, 2024, 6:43 p.m.