Requirements

When the manuscript was written, Guide was available only for the Mac, but now it works on all platforms where R can be run, hence should work across Mac, Windows and Unix systems. Guide now works as an R package, hence its main requirements are that R is installed, as well as an R packages opencpu, limma and edgeR.

Guide runs on a browser, and it should run on most modern (IE10+) browsers capable of running HTML5.

For more detailed instructions on installing the required programs and packages, follow the steps below.

1. Install R
R can be obtained from r-project.org, (look for download R link). It needs to be installed as framework (this is the default option when installing R). Guide has been tested with R version > 3.0.1.

2. Install opencpu, limma and edgeR
R package called opencpu needs to be installed. You can do this by starting R, and typing in the following command:
    install.packages("opencpu")

Guide also relies heavily on R's limma package, and requires edgeR package for RNA-seq data processing. This can be done from within R by using the following commands:
    source("http://bioconductor.org/biocLite.R")
    biocLite("limma")
    biocLite("edgeR")

For Mac OS > 10.8 users:
Guide requires X11 to be installed, and X11 is no longer included with the operating system from OS 10.8. X11 can be obtained from xquartz page.

Installing and Running Guide

1. Download the latest version: guide_1.2.tar.gz.

2. Start your R session, and install Guide as R package where /path/to/download/ here is the example path to where the downloaded package is located:
    install.packages("/path/to/download/guide_1.2.tar.gz", repos=NULL, type="source")

3. You're now ready to run Guide. Still in your R session:
    library(opencpu)
    ocpu_start_app(‘guide’)
 
(Note this last command has changed with the latest version of opencpu (2.0+). So you may be running an older version of opencpu if the last command fails. Upgrade opencpu by just installing it again, as in step 2 above.)

This will open up a new tab on your default web browser and start your Guide session..

Troubleshooting

If you encounter an error, you can try contacting the developers. Using the javascript console tool on your browser to show any error messages and sending these is also helpful. Due to limited resources, Guide is not undergoing any further development, but bug fixes do happen.

Guide has been written in R, javascript and html.

Release Notes

version 1.2 (2016-11-21):

  • Fixed numerous bugs, including one where some datasets didn't upload properly.
  • save gene set function which was buggy has been replaced by the more reliable export gene set function.
  • Updated several javascript libraries to be more up to date - though the end-user may not experience any direct difference due to these updates.

version 1.1 (2015-02-12):

  • This is a completely new version in terms of its architecture. All previous versions were stand-alone applications built in Qt. Now the application is built primarily in javascript, and provided as an R package, rather than as a desktop application. This change has been made for several key reasons:
    1. Installation of R as a pre-requisite is now easy to enforce, as the program runs from R.
    2. It is no longer necessary to provide separate versions of the program across different platforms (it was much more difficult to create windows and linux versions in Qt than anticipated).
    3. Previous versions also depended a lot more on the OS and R versions. This is now greatly simplified.
    4. Web interface is more intuitive for most users and can use latest technologies such as d3.
  • As a result of this major change, this version serves as a prototype. Hence only minimal features available from the previous versions have been added for now.

version 0.3.1 (2014-06-05):

  • Fixed a bug where pathway analysis wasn't working for limma version > 3.16.0. Note that this version of Guide now requires that limma version > 3.16.0 be installed for the pathway analysis to work.
  • Fixed a bug where heatmap wasn't rendering for microarray data.
  • Added gene description column in the file which is saved when a gene set is saved.

version 0.3.0 (2014-02-04):

  • Fixed a bug where MDS plot was giving an error message with microarray data.
  • Fixed a "bug" where if a gene had multiple probes with significant p-values, the score assigned for the gene is now based on the probe with lowest adjusted p-value, rather than hightest.
  • Fixed a bug which prevented new sample group definitions showing up on the Dataset Analysis page.
  • Not having Rcpp installed before running Guide should not crash Guide now. There will be a warning message and an option to try automatic installation will be shown.
  • If edgeR or limma isn't installed before trying to run heatmap, you should now get a warning message and an option to try to install these automatically.
  • Fixed a bug where a warning was shown after running differential expression analysis using non-default adjusted p value (eg. setting this to 1).

version 0.2.9 (2013-09-27):

  • Guide now provides an installation package, which should make it slightly easier to install it. You still need to have R and Rcpp installed first, but running the installation package will now create a "Guide" folder in your home directory, and inside it you will find the application as well as data files.
  • Fixed a bug which made Guide unresponsive when trying to save a geneset while on Mac OS 10.6. (Note that instead of file dialog, you'll get a message indicating that your table has been copied to the clipboard, if you're running this OS.)
  • New menu items under the Tools menu show how to upload unsupported microarray data, as well as adding data for unsupported species.
  • You can now plot a heatmap for any gene set. Future versions will refine this feature.
  • Plotting can now handle errors better, by displaying the error messages rather than being silent.

version 0.2.8 (2013-08-23):

  • Guide works now for R version 3.x
  • Fixed a bug where sometimes plotting on the dataset did not show the plot properly.
  • RScripts/topTable.r and RScripts/topTableMicroarray.r have been modified to not throw an error if non-character vector is passed on as group variable (this shouldn't happen within Guide, but if using the script directly from R, it could happen).
  • Created an application icon for the mac.

version 0.2.7 (2013-05-08):

  • Not having R or Rcpp installed on the machine won't crash the program now. It performs a check, and shows a suitable page if this is the case instead.
  • Fixed a bug where editing the R code on the Data Analysis page didn't work.

version 0.2.6 (2013-04-18):

  • Affymetrix MG-430 PM Array is now supported.
  • After a new sample group is defined for a dataset, the table of sample groups on the differential expression page now refreshes properly.
  • Added new columns to Enriched GO Pathways Analysis page, to show the number of common genes between the pathway and the geneset tested. Clicking on either column will show the common genes.
  • Added some explanatory text to camera function within pathway analysis.
  • Status bar text resets when switching pages.
  • List of sample group entries are now alphabetically sorted on the Differential Expression Analysis page.
  • The pages showing user text file input (dataset upload and sample groups file upload) have some text explaning that 'windows formatted text' must be used when saving files from Excel on a Mac.
  • Fixed a bug where clicking on the "type" column of the Uploaded Datasets page crashed the program.

version 0.2.5 (2013-03-19):

  • Fixed some outdated help text on Differential Expression Analysis page.
  • "View Pathway Analysis Results" menu item under the Genes menu now shows the correct pathway analysis page, and "View Enriched GO Pathways Results" menu item has been added.
  • Added chromosome annotation to each gene.
  • Improved load gene set function on the gene set page, so that a score column position can be specified.
  • Differential Expression Analysis page now automatically selects a different sample group for the second sample group.
  • Some more help text has been added to Differential Expression Analysis page.
  • You can now input your own number of minimum expressed samples when filtering for low count genes on the Differential Expression Analysis page.
  • Added some help text on the status bar for gene set page.
  • Search field on the Search by term page now has the focus as soon as the page is shown.

version 0.2.4 (2013-03-04):

  • Significantly changed the way enriched GO terms are worked out, by using data from Entrez rather than Broad Institute (which had the issue that mouse gene sets were derived using orthologues rather than directly from GO).
  • Also enriched GO terms are split based on categories.
  • Improved new dataset upload function where the dataset has only Ensembl ids.
  • Differential expression search options are now called "differential expression" and "pathway analysis" instead of "limma/voom" and "camera". There is no change in the way that calculations are performed.
  • If limma package is not installed when trying to upload raw microarray data, it will now try installing it.
  • Help menu item now goes to the documentation website.
  • Fixed a bug where changing dataset name wasn't followed by other pages.
  • Added "Pickrell" dataset as an example dataset.
  • Fixed a bug where "View Pathway Analysis Results" menu item was showing the wrong page.
  • Added a version check at the start of the program.
version 0.2.3 (2013-02-25):
  • First version to be released