Since almost 2 years I am working for a global Bio pharmaceutical company. Together with researchers, I am working on applications to analyze (sometimes big) data to find medicines for patients with serious and life-threatening diseases. As you can imagine, the work is confidential and so I am not able to share data or results with other people. However, there are a couple of initiatives to open-source some of the packages that are developed within the company, which is awesome! The Shiny contest gave me a good opportunity to display some of these packages in a Shiny application.
Periscope is a new R package which provides a predefined but flexible template for new Shiny applications with a default dashboard layout. Furthermore it provides user alerts, a nice busy indicator, application reset and logging features.
One of the most important features of the Shiny applications created with this framework is the separation by file of functionality that exists in one of the three Shiny scopes. These scopes are: global, server-global, and server-local. The framework forces application developers to consciously consider scoping in Shiny applications by making scoping distinctions very clear without interfering with normal application development. Scoping consideration is important for performance and scaling. This is critical when working with large datasets and/or across many users. In addition to providing a template application, the framework also contains a number of convenient modules.
- (multi)file download button module
- downloadable table module.
The plots in the application are created using CanvasXpress, which is a package for (interactive) data visualization especially for reproducible research. It supports a large number of visualizations to display scientific and non-scientific data which includes: Area, AreaLine, Bar, BarLine, Boxplot, Bubble, Candlestick, Chord, Circular, Contour, Correlation, Density, Donnut, DotLine, Dotplot, Genome, Heatmap, Histogram, Kaplan-Meier, Layout, Line, Map, Network, NonLinear-Fit, Oncoprint, ParallelCoordinates, Pie, Radar, Remote-Graphs, Sankey, Scatter2D, Scatter3D, ScatterBubble2D, Stacked, StackedLine, StackedPercent, StackedPercentLine, Sunburst, TagCloud, Tree, Treemap, Venn, Video, Violin. This blogpost gives a good overview on how to get started with CanvasXpress.