[Bioc-devel] AnalysisPageServer: A framework for sharing interactive data and plots from R through the web.

Brad Friedman friedman.brad at gene.com
Sun Dec 21 09:06:05 CET 2014


Dear Bioc-Devel:

I'm pleased to present our package AnalysisPageServer
<http://bioconductor.org/packages/devel/bioc/html/AnalysisPageServer.html>,
now available in the devel branch of bioconductor.

AnalysisPageServer is a framework for sharing interactive data and plots
through the web. Using AnalysisPageServer, one can embed interactive plots
and data tables in otherwise static reports such as knitr documents,
delivering the following features:

   - Rollover of plot elements to show meta-data
   - Filtering of table and plot based on meta-data
   - Selecting and tagging of plot elements with filtering based on
   selection
   - Download of (filtered or unfiltered) data in .csv, or plot in .svg or
   .png.

Screen shots of these features
<http://bioconductor.org/packages/devel/bioc/vignettes/AnalysisPageServer/inst/doc/Interactivity.html>
are available, and we also demonstrate a fully interactive vignette
<http://www.bioconductor.org/packages/devel/bioc/vignettes/AnalysisPageServer/inst/doc/embedding.html>.
Indeed, AnalysisPageServer makes it easy to include these UI features in
any Bioconductor vignette. These reports would also be an excellent format
for supplemental data for academic journal articles.

AnalysisPageServer can also be used to turn R functions into full-featured
web applications deployed with Rook, FastRWeb, or RApache. Function
arguments are rendered as form input widgets, then custom plots and data
are generated upon user request and sent to the front end for further
interactive browsing and analysis. R programmers can focus on data logic
and let AnalysisPageServer handle all the details of the user interface.

Compared to other popular systems for building web applications in R,
AnalysisPageServer offers a great deal of interactivity, giving users more
power to explore their data. Furthermore, all features are included
completely free and ready for enterprise-scale deployment. Using this
system we have deployed a web application in a single VM with 10GB memory
which serves approximately 1500 plots and datasets for our research
collaborators each week.

AnalysisPageServer is a project of the Genentech Bioinformatics Department.
I hope you will find it useful and would be happy to answer any questions.

Sincerely,

Brad Friedman

	[[alternative HTML version deleted]]



More information about the Bioc-devel mailing list