Writing custom panels for the iSEE package

The iSEE package provides an interactive user interface for exploring data stored in SummarizedExperiment objects (Rue-Albrecht et al. (2018)). This repository hosts the source code for minimal iSEE applications that each demonstrate a custom plot or table panel for the iSEE package.

Custom plot and table panels are described in the vignettes of the iSEE package. Briefly, custom panels allow users to add an arbitrary number of functions that process a SummarizedExperiment object, a selection of rows, and a selection of columns to produce a ggplot object or a data.frame from dynamically computed data, unlike predefined plot and table panels.

Repository organization

Each example is stored in a separate subfolder. Folder names should start with table_ or plot_, to indicate the type of custom panel and facilitate browsing.

Each example must be comprised of five files:

• custom.R: a script that defines the function(s) underlying the custom panel.
• app.R: a script that prepares a small data set, configures the iSEE application, and launches the tour.
• tour.txt: a set of step-wise instructions attached to various UI elements in the iSEE user interface.
• Screenshot.png: a screen capture or illustration of the custom panel that will be shown as a thumbnail in this README file. The image should not include more than 1 row of 3 panels, to be displayed in a width="450px" height="150px" format.
• README.md: a file that displays Screenshot.png in a width="100%" format.

To launch an application, simply set your working directory to the appropriate subdirectory, and execute app.R.

Examples available

Click on the the image to access the source code.

Screenshot	Description
	Table of log fold-change with cache. Compute the log fold-change between a selection of samples and all other samples. Restrict the result table to a selection of features. Cache log fold-change values for all features. Only recompute them when the selection of samples changes. Changing the selection of features simply restrict which rows of the cached results are displayed.
	Multiple reduced dimension plots. Visualize the distribution of a selection of samples across all dimensionality reduction results. Changing the selection of samples simply highlights those samples in all the plot panels.
	Read coverage plot. Visualize the read coverage of a genomic region, as well as a volcano plot. Changing the selection of a gene in the volcano plot (or providing a gene ID directly in the custom coverage plot) shows the read coverage in the region of the selected gene.