Fixed Thread
Form http://archive.ambermd.org/201209/0203.html
A chat format
Jonathan Gough:
The basic (or complex) question I have is:
How do you take a PDB and change one residue to another residue?
(essentially a point mutation of an existing structure)I thought I remembered reading how it could be done, but looking back I can’t seem to find where it might be. I can think of a few ways one could accomplish this, but I wanted to ask if there is an explanation in the manual or a tutorial that I am just missing. (Can someone point me in the right direction)
I searched the archives and saw some old posts regarding using other programs.
One could manually add/change things
3+????Not necessarily looking for a step by step but a push in the right direction.
Francois:
You could edit the PDB file: (i) remove the side chain of the amino acid to be mutated; (ii) rename the backbone of this residue according to the residue name of the mutation. Then, you load the modified PDB file in the LEAP program, which will automatically add the missing atoms (i.e. the side chain) in agreement with the FF library of the mutated residue.
Carlos Simmerling:
This isn’t a good method- leap doesn’t care at all where it puts the side chain and unless you’re very lucky you will have bad steric clashes that can invert chivalry and other bad things. You want to use a program that searches rotamers for something that fits as best as possible
Carmenza Martinez:
Not sure if anybody has suggested it previously but for single point mutations or multiple mutations of residues I have found swissPDB to be quite useful: http://spdbv.vital-it.ch/ Just as Prof. Simmerling leap is not efficient at filling in the blanks when you remove things and renamed them just as Francois suggested. Now, when you say 3+???? I don’t understand what you mean…changing the protonation state perhaps? for that you will need more than what swiss PDB could provide. Perhaps somebody else in the forum will provide useful advide for that.
sorry I meant to say just as Prof. Simmerling said…the said got deleted…sorry
Jonathan Gough:
Thank you all for your help! Very good suggestions. I am using swiss PDB right now.
I said 3+ as I could think of at least 1 if not multiple more ways to do it (other applications or combinations of applications)
Carlos Simmerling:
what often happens in my experience is that the steric clashes are bad enough that the atoms get pushed around, and the chirality inverts because of that. eventually you minimize the clash away, but in the process things are no longer correct. we’ve even seen really weird cases such as where a Phe ended up with a protein chain going through the middle of the ring obviously no way that will ever get fixed in MD.
I guess I’m just warning people that if you have a high energy structure, just because you minimize it doesn’t mean things are ok. you should always visually inspect the area where you made the change.
Aron Broom:
Yikes, point taken.
Also for anyone who finds this thread, PyMol has a mutation function that uses a rotamer library
Francois(In a different context):
You follow the same approach for a modified nucleotide residue than
for a mutated amino acid residue. You remove the base and rename the
residue name for the ribose derivative according to the residue name
defined in the FF library for the mutated/modified nucleotide. x/tLEaP
will do the job and add the missing atoms for the modified residue.To convince Carlos (if one needs to) LEaP has a geometry optimizer
(all structure or selected parts), can invert chirality center and
modify dihedral angle values to correct the generated modification.
Most of these commands were only available in xLEaP; at
q4md-forcefieldtools.org we have introduced them in tLEaP allowing to
make these commands ‘scriptable’…
See http://q4md-forcefieldtools.org/Tutorial/Tutorial-1.php
http://q4md-forcefieldtools.org/Tutorial/leap-mol2.phpThe strength of LEaP (I mean x/t) is that it is free (sense of
freedom), it is directly usable (when one has installed the
AmberTools) and very powerful (when one uses it the correct way);
obviously one needs to understand how it works (may be not that easy).
The Email format
The is reformed emails are in the folded block.
|