[gmx-users] ihedral restraints in v3.3 different than dihedral restraints in v3.3 different than v3.3 (or broken)

[chris.neale at utoronto.ca]

> I try to perform dihedral_restraints on protein helices. Searching   
> on this subject in the gromacs manual, wiki.gromacs and gmx-mailing   
> list give me some clue but it is not clear for me yet. Also I did   
> some you suggest in following address:
> http://www.gromacs.org/pipermail/gmx-users/2006-December/025087.html
>
> In my top file I added lines below and include all the atoms in the   
> helices in case of A,B,C,D:
>
> [ dihedral_restraints ]
> ; ai   aj    ak    al  type  label  phi  psi   dphi  kfac  power
>    A    B     C     D    1     1    180  180     0     1      2
>

If you take a closer look at the information that I posted in the link  
that you referenced, you will notice that I do not have "psi" listed.  
You need one entry per dihedral.

You will also notice that my comments were in reply to somebody asking  
about "bonded atoms A-B-C-D" and so they were just placeholders. You  
really want the number of that atom as you find it in the .itp or .top  
file (not the .gro file).

So, assuming that you have atoms numbered like this in your peptide.itp

C' (n-1) = 5
N  (n)   = 7
CA (n)   = 9
C' (n)   = 15
N  (n+1) = 17

Then you would use a section like this:

#include "peptide.itp"
[ dihedral_restraints ]
; ai   aj    ak    al  type  label  phi  dphi  kfac  power
; phi C'(n-1) - N - CA - C'
    5    7     9    15     1      1  180     0     1      2
; psi N - CA - C' - N(n+1)
    7    9    15    17     1      1  180     0     1      2

#include "tip3p.itp"
; Note that the order was 1) peptide .tip, 2) peptide dihedral  
restraints, 3) other stuff...


> And in the mdp file I include these lines:
> ;dihedral restraints
> dihre               =  simple
> dihre_fc            =  100
> dihre_tau           =  0.0
> nstdihreout         =  50
>
>
> But I have got error in processing the top file.
>
> Fatal error:
> Ther were 1 error(s) processing your input

I happen to have known what was going wrong here, but it is always  
better if you post the whole output for an error (assuming it isn't  
thousands of lines). If you had pasted right from your grompp command  
to the end of the output, I assume that there would have been more  
information that one could use to help you. So if this did not solve  
you problem, or if you have another problem in the future, then please  
post more of the output.

Chris.

>
> I have referred to the manual and not found any thing about   
> dihedral_restraints details. Would you please explain to me what I   
> should do.
> Thank you very much in advance for your kindness and expertize.