Dynamic multivariatE data analysis And VIsUalization PlatforM

A GUI for interactive plotting and analysis of multivariate data.

Including:

• GUI using RGtk2 toolkit implemented with gWidgests

• Dynamic plotting - base and ggplot2

• Linked brushing of multiple plots - iplots

• Network Visualizations - igraph (static, interactive and 3D), Cytoscape (RCytoscape)

• Analyses - univariate and multivariate

• Automated report generation - rsweave

• Data import and linking with MS Excel and Google Spreadsheets

• Successor to imDEV, including improved interface and capabilities

Eventually devium will be installed from R as a package and from alternative repositories where it can be bundled with GTK+ components. For now it can be installed by using the source code in the devium/R directory. Copy and paste the following code in to R to source the necessary files.

#load package from github repo
source("http://pastebin.com/raw.php?i=JVyTrYRD")

#check out some of the functions (accesory objects will be cleaned up)
objects()

#launch GUI (watch R console for messages regarding the downloading of dependancies)
devium.gui()

Note if you don't yet have the GTK+ toolkit installed you need to specify to "Install GTK+" when prompted from the R command line.

Prototypes for some of the analyses implemented in Devium can be viewed in the browser using the Shiny frame work. Check out some of the examples below.

• Heirarchical Cluster Analysis
• [Basic Plotting] (http://spark.rstudio.com/dgrapov/1Dplots/)
• [Principal Components Analysis Diagnostics] (http://spark.rstudio.com/dgrapov/PCA/)