How to Add a Symmetry Related Molecule using Coot
You can do this by using Coot and a simple text editor. This can be helpful in preparing figures that discuss biological function. The steps are as follows:
1) Display the symmetry related molecules:
Draw -> Cell & Symmetry…
Note: I changed the master switch to ‘Yes’, adjusted the radius to 50, and then hit ‘Apply’
Below is an overview of the crystal packing:
2) Attain a perspective that will allow you to easily click on your desired symmetry molecule. Below is a zoomed in view of that same packing.
3) File -> Save Symmetry Coordinates
Your pointer will become a small cross shape. Now, click on your desired symmetry molecule.
4) Give the file a name you can remember for at least 15 seconds:
5) Open your original pdb file and then newly generated symmetry pdb files in a simple text editor. I am using gedit on a linux machine in this example.
6) Rename the newly created pdb file from ‘chain A’ to ‘chain B’.
Note: If you do not rename the chains, the atoms may create strange connectivity such as long distance disulphide bonds in cysteines.
Important: I have an ” A ” (there is a space on either side of the A). I always scroll through the pdb file to make sure I’m not replacing the letter A in an undesired place. For example, if you simply replace “A” you will mess up all your CA (carbon alphas) and residues abbreviations such as VAL.
7) Copy and paste ‘chain B’ into your original pdb file. You should copy/paste the ‘ATOM’ section and leave out the header (everything above the ATOM section) and the END (it is at the bottom of your pdb file).
Rename the file and save using ‘Save As’. This way you will have the original monomer available should you need it in the future. If you accidentally overwrite the original, you can always just delete the second chain and re-save the file.
9) Display your newly created pdb file and make sure that your new pdb file is correct.
wangsa
September 8th, 2009 at 9:11 AM #
Many thanks ! I tried PDBset but failed and this one worked perfectly
Sean
September 9th, 2009 at 6:35 PM #
Awesome! Glad to hear and thanks for taking the time to drop me a note.
Saif
September 17th, 2009 at 3:49 PM #
At step 6, when you rename the chain in the new file, do you do it manually ? Is there some way other than doing it for each atom since my protein has over 2000 amino acids.
At step 7, when you copy and paste your new chain, do you leave out the end section with the “CONECT” and other columns ? Where exactly do you paste the new chain- right after the previous chain ends and before its “CONECT” section starts ?
Thanks a lot.
Sean
September 17th, 2009 at 4:11 PM #
Hi Saif,
At step 6, I am using the find and ‘Replace All’ (see the picture) with ‘ A ‘ chain with that of ‘ B ‘ chain (use a space on either side of the letter to only select your desired chains and not accidentally replace the carbon alphas (CA)). Before clicking ‘replacing all’, it is worth a quick look through the file to make sure you are replacing what you want (you should be able to highlight them for easier viewing). I am renaming the whole chain at once so it is very fast.
At step 7, I paste below the “previous” chain without the CONECT (assuming you are copying chain B to put below chain A). I have never had an issue leaving out CONECT at the end of PDB files. However, I do make a backup copy just in case.
Hope that helps and good questions!
Paul Emsley
February 21st, 2010 at 4:22 PM #
I missed this earlier.
I need to add a function to generate a symmetry molecule without writing to disk first. Hmm…
Anyway, watch out for the imminent PISA interfaces and assemblies wrapper in Coot, which should do much of what you describe above (and represent interaction surfaces and hydrogen bonds).