GEP header

How to Add Reads to a Project

After you've received the new reads you ordered, you will need to incorporate them into your project.

  1. We can't phredPhrap the new reads into our projects on USB drives, we have to do this inside virtual hard drives. The least troublesome way to do this is to download your new reads package from the GEP Data Repository (on the GEP website under the "Projects" tab) directly to your virtual hard drive or your USB drive (and from there to the VHD); unfortunately we're having some firewall problems which means this isn't practical for another couple weeks (i.e. until we've completed the finishing portion of the course). So I've downloaded the new read files to my USB drive at home and will provide them to you in class.
  2. When you start up your virtual machine (assuming you've enabled a virtual hard drive with it), when the "boot:" prompt comes up (on that initial olive-drab screen with the GEP banner across the top), type slitaz home=sda1 [note the space between "slitaz" and "home=sda1"] and hit Return — this will make the virtual hard drive the default home directory (home > tux).
  3. (The first time you do this, you'll need to copy your current project to the virtual hard drive: navigate to the directory containing your project folder/directory and issue the command tar -czvf [name of file to be created] [project name]. Lots of file names will scroll by — this can take some time, depending on the size of your project directory — it's done when you see the prompt again. To copy this compressed folder to your home directory (tux), issue the command cp [name of compressed file] /home/tux/. Navigate to your home directory and decompress your project file by issuing the command tar -zxvf [compressed file name].)
  4. Mount the USB drive containing the compressed (.tgz) file containing your new reads.
  5. Copy your file (it will begin with "11round" and have your project number in the name) to your home directory.
  6. Decompress your new read file by issuing the command tar -zxvf [compressed file name].
  7. You should now have four items in your home directory: your project directory, your new reads directory, and compressed versions of both of these. Ultimately you can get rid of the compressed versions by issuing the command rm [file name], but for now I'd leave them (until you're sure they're not needed).
  8. Move the .fof file from the new reads directory to the edit_dir directory of your project by issuing the command mv [name of the .fof file] ../DMAC_##a/edit_dir.
  9. Move all the remaining files in your new reads directory to the chromat_dir of your project by issuing the command mv [name of the file] ../DMAC_##a/chromat_dir. Note that you can use the wildcard symbol "*" to move many files at one time: using "*.b1" as the file name will move every file that ends in ".b1" in one fell swoop.
  10. There are two ways to incorporate your new reads. The preferred method is to re-phrap the entire project, starting from scratch but with the addition of your new reads. The downside to this approach is that if you have corrected a missassembly in Consed (by tearing a contig or forcing a join, for example), phrap will probably make the same mistake again, forcing you to correct the mistake(s) every time you re-phrap. To re-phrap your entire project, simply redirect the virtual machine to your project's edit_dir by issuing the command "cd ../DMAC_##a/edit_dir" followed by "phredPhrap".
  11. If you don't want to re-phrap the entire project, you can add the reads to the existing project, but before you can do this you'll need to edit a perl script in Consed; you can open Consed now or later, your choice, but you have to do this before adding reads. Thanks to Jon Rekai for figuring this out and providing the instructions (which may not work with this version of things, I haven't tested it in 2011 yet):
  12. In your virtual machine (not Consed), double-click the Apps folder. (Note that right-clicking on the virtual machine's desktop and choosing Apps in the cascading menu won't work for this.)
  13. Double-click Root. A terminal with blue text and a red cursor opens.
  14. Issue the command "vi /usr/local/genome/bin/determineReadTypes.perl".
  15. Issue the command ":180" — this navigates to the offending line in the script.
  16. Issue the command "dd" — this deletes that problematic line of code.
  17. Issue the command ":w!" — this saves that change.
  18. Issue the command ":q" to exit the editor.
  19. Hit the Control key and "d" to exit the terminal. (Note that these changes won't persist across reboots of the virtual machine, so you'll have to do this every time you want to add reads this way. Once they're added to your project and you save the new version, though, you won't need to add those reads again in future.)
  20. Go back to Consed, and click the "Add Reads" button in the Main Window.