[gmx-users] pdb2gmx... cys-cys bonds....

[Mark Abraham]

sharada wrote:
> Dear  Dr. Mark Abraham,
> 
> Thank you for the response. I am into desiging small antibacterial 
> peptides with cysteins. I would have to play around with forming 
> different combination of disulphide bonds and check the stereochemical 
> properties and further design them experimentally to see if they are 
> active. I have many peptides  to build models of a lot of  structures 
> with different Cys-Cys combinations  for my simulations as my starting 
> structures.  If I want to form only those cys-cys pairs then yes I would 
> have done the MD with harmonic restraints and bring it near.   I have to 
> check  the proximal cys-cys distances and then make a decision on 
> building only those disulphide bonds whose S-S distance are  in the 
> range of 4A to 5A.    Can I change the pdb2gmx program to make this 
> decision?   Is it  possible to change the code?

You don't really want to change pdb2gmx. You want to write a Perl script 
that automates the disulfide bond selection process, does the harmonic 
restraint MD, and then simulates using a vaguely sensible disulfide 
starting structure.

>  if so how do I 
> incorporate these changes. I think the pdb2gmx program is checking for 
> S-S distance to less than or equal to 2A and makes the bond . Am I right 
> ? 

You can edit specbond.dat yourself... the format isn't described in the 
manual, if I recall correctly, but it's pretty obvious... you just 
increase a 0.2 to 0.4 or some such.

Mark