This file is for admins/support-staff who want to configure and install the PTXQC package and provide an automated drag'n'drop solution to their MaxQuant users (incl. themselves) for creating QC reports. Please contact an admin or proficient R user, if you are new to installing R packages.
Note: the drag'n'drop currently only works for Windows, not MacOSX or Linux. For the latter, please refer to the vignettes of this package (after installation), which explain how to invoke PTXQC from within R.
We recommend using a shared network drive as installation folder to which all users have read access. Alternatively you can copy the folder structure we are about to create to a folder on your local machine. However, if you want to use the QC reporting from multiple PC's just use a network folder.
When you are done, provide the users with the PTXQC-DragNDrop vignette (or write your own). This will show them how to invoke the QC (it's very easy - see 'Usage' below).
In order to make PTXQC available as a standalone application on a network drive, we will need to make a copy of the whole R installation directory as one of the last steps. If you use an existing R installation which already contains a large package library, you might want to find a fresh PC without R and install R from scratch, such that you do not end up copying useless packages which are not relevant for PTXQC. If you are a knowledgeable R user already, you can probably find other ways around it (e.g. a temporary library, or selective copying of packages, but this is out of scope of this manual).
- create a clean R installation by downloading R from [http://www.r-project.org] and installing it. We highly recommend the 64bit version to avoid out-of-memory errors. If you already have R installed, you may skip this step.
- install pandoc (see bottom of linked page). Pandoc is required in order to locally build the package vignettes (documentation), but you can also read the vignettes online from the PTXQC GitHub page. More importantly, Pandoc enables PTXQC to write QC reports in HTML format (which come with a help text for each plot and are interactive). PDF reports only contain plots! The reports are printed as PDF by default and additionally as HTML if Pandoc is found. If you install Pandoc later while your R session is already open, you need to close and re-open R in order to make R aware of Pandoc!
- Now, we install PTXQC and its dependency packages.
Start the R GUI (64bit!) and execute the following commands to install the
PTXQCpackage (dependencies are installed automatically).
You can grab PTXQC from either CRAN or GitHub. GitHub installation will give you the latest package; the CRAN version might be a little older, but is faster to install. Check the [NEWS][News_File] file for CRAN submissions and version.
For the code blocks below: Run each line separately in your R console, i.e. do not copy and paste the whole block. If an error should occur, this allows to track it down more easily. See FAQ - Installation how to resolve them.
## CRAN install.packages("PTXQC")
## GitHub if (!require(devtools, quietly = TRUE)) install.packages("devtools") library("devtools") ## this might give a warning like 'WARNING: Rtools is required ...'. Ignore it. ## use build_vignettes = FALSE if you did not install pandoc or if you encounter errors when building vignettes (e.g. PRIDE ftp unavailable)! install_github("cbielow/PTXQC", build_vignettes = TRUE, dependencies = TRUE)
Finally, always run:
cat(paste0("\nPTXQC was installed to '", .libPaths(), "'.\n\n"))
Here's some intuition about the next steps
To make PTXQC Drag'n'drop a standalone application which can be run from any user and even from a network drive, we need a copy of the whole R installation. Within the PTXQC package (which is part of the R installation) we have prepared a drag'n'drop batch script which invokes R and calls PTXQC. We need to extract this script (i.e. copy it to a top-level folder) and then place the whole R installation beneath it. This includes all your R-libraries, which might reside in a different place.
Almost done, we just need to copy some folders.
- The last command will tell you where R installed PTXQC, i.e.
PTXQC was installed to '<libR>'.Open this
<libR>folder in your file explorer. It will either be your default R library folder which comes with R (e.g.
C:\Program Files\R\R-3.1.0\library), or a user-specific library folder like this
C:/Users/cbielow/Documents/R/win-library/3.1. Keep this directory in mind. We will need it later.
- Copy the folder
<libR>\PTXQC\dragNdrop\QC-dragdropto a custom target location of your choice (let's call this target folder
<QCdir>) where you want PTXQC to reside. Usually, that's some network or local drive (e.g.
Z:\my-proteomics). Now you should have
- Copy the whole R installation directory (e.g.
c:\program files\R\R-3.1.0) into the
<QCdir>\QC-dragdrop\_internalsub-folder, such that you end up with
<QCdir>\QC-dragdrop\_internal\R-3.1.0\(your R version number might differ)
- Copy the Pandoc.exe (which you installed earlier) to
<QCdir>\QC-dragdrop\_internal\R-3.1.0\bin\pandoc\pandoc.exe(you will need to create the pandoc folder). The batch file expects to find Pandoc there and will throw an error if Pandoc is missing.
- If your newly copied R installation directory is not named
R-3.1.0, rename it to
R-3.1.0(or edit the
<QCdir>\QC-dragdrop\createQC_dragNdrop.batto match your R version)
- If the
<libR>folder was not your default R library (see step 1), but a user library, then copy the packages it contains (PTXQC among others), into
Now, you should have the following structure
QCdir\QC-dragdrop\ \createQC_dragNdrop.bat \createQC_dragNdrop_withYAML.bat \_internal\ \R-3.1.0\... (version may differ) \compute_QC_report.R
You can rename the
QC-dragdrop folder to anything you like (try to avoid spaces - they usually cause trouble).
To make PTXQC accessible more easily, you can create a shortcut on your Windows Desktop pointing to
createQC_dragNdrop.bat. Use this shortcut as if it were the original batch file (see usage below).
Updating the PTXQC Drag'n'drop installation
Start the R GUI within in PTXQC standalone installation at
run the block of code from the installation (see above) again. Then, copy the packages from the
<libR> folder to the main R library (see the last bullet point above).
Important: make sure that you start with a clean R-Session and that no other R-Sessions are open, since they might prevent the update of package files where
R will only give you warnings like
Warning: cannot remove prior installation of package "digest", but will fail hard with a misleading error message afterwards.
Usage (short version)
For the long version, see the PTXQC-DragNDrop vignette of this package.
You can create a QC report by dragging a txt-folder (or any file within the txt-folder)
createQC_dragNdrop.bat file which resides on your (network) drive.