alphafold on rhino (Campbell Lab Implementation)

Instructions (summary):

1. Log into RHINO cluster
2. Go to the directory where you want to run alphafold
3. Put your fastafile in that directory
4. Copy this script into that directory: /fh/fast/campbell_m/share_scripts/share_run_alphafold_cluster_v1.0
5. Run the program using ./share_run_alphafold_cluster_v1.0

Example:

ssh melody@rhino
cd /fh/fast/campbell_m/user/melody/alphafold
vi integrin.fasta
cp /fh/fast/campbell_m/share_scripts/share_run_alphafold_cluster_v1.0 .
./share_run_alphafold_cluster_v1.0

*Anywhere you see pink text, substitute with your own info
**Do not substitute your own info in the path to the actual program

Notes on how to copy the script:
- Go to the directory where you want to put the script, then use this command: cp /fh/fast/campbell_m/share_scripts/share_run_alphafold_cluster_v1.0 .

Notes on uploading, transferring, or creating files:
- You can import a fast file by either using a command like rsync, scp, or wget or making a text file (using vi or nano or your fav text editor) and copy pasting the sequence in. Alternatively, if you're familar with it you can use Motuz.
- Make sure to keep the title with the > as the first line of the fasta file

• Uploading a file via terminal example (These directions will work on a mac)
  • Put your .fasta file on your desktop (NAMEOFFASTA.fasta)
  • Open a terminal
  • Use this command to get you home cd
  • Move to your desktop cd Desktop
  • rsync -rtoulv example.fasta melody@rhino:/fh/fast/campbell_m/user/melody/.
    • anything in pink needs to be substituted with your own info
      
      • sub in your .fasta file name, your rhino username at each instance of "melody", as well as your PI name for the "campbell_m"
      • your rhino username is probably more like mcampbell, jhollis, or anguyen4. it is unlikly to be your first name.
        
    • you can also put your fasta wherever you want, above is just an example
• Uploading a file using Moutz
  • Fred Hutch documentation about file systems and Motuz: https://sciwiki.fredhutch.org/scicomputing/store_posix/
  • Link to Motuz: https://motuz.fredhutch.org/
• Transfering a file using wget
  • To use this method, fasta file must be hosted somewhere online
  • Example:
    • Search for your protein on UniProt 
    • Once you find it, click on the "Entry" (alpha numeric)
    • On the left column click on "Sequence" (or "Sequence & Isoform")
    • Click on "Download"
    • This will open a new window with the contents of the fasta file
    • Copy the URL
    • Go to your terminal
    • Bring over the file using wget
      • Example: wget https://rest.uniprot.org/uniprotkb/P11215.fasta .
    • Rename if desired
      • Example: mv P11215.fasta ITGAM_human.fasta                 
• Creating a file using vi
  • (have your fasta sequence copied via command + c)
  • login to terminal and go to the directory where you want to put the file and run alphafold
  • vi example.fasta (in terminal. this will open a new blank file with the name you’ve given it)
  • i (just press the i key, it will put you into insert mode)
    command + v (this should paste your sequence, you should see it in the window)
  • esc (this gets you out of insert mode)
  • :wq (this saves the file and quits vi)
  • to check, you can use this command which should display the file: cat example.fasta

Notes on logging into the cluster:

• If you have never logged into rhino before you need to set up an account. To do this, email scicomp@fredhutch.org. 
• Likewise, if you cannot log into the Fred Hutch Rhino cluster, please direct all of your questions to scicomp. 
• Here is more info about logging into rhino: https://sciwiki.fredhutch.org/compdemos/first_rhino/

If you want to learn a bit more about command line, you can watch this video, starting at 11:39 to go over some extremely basic commands. (Please excuse my vocal fry; it's just my voice.)