[gmx-users] Purpose of repeated improper dihedrals

[Justin Lemkul]

On 8/2/19 8:59 AM, Dawid das wrote:
> Thank you for an asnwer. However, I still need to ask more.
> So, let's take this time the
> NR1  CPH1  CPH2  H
> NR1  CPH2  CPH1  H
>
> example. For such quartets,  the improper dihedral parameters are defined
> for HSD residue.
> Now, the bonding is as follows in part of HSD:
>            H
>            |
>          NR1
>        /        \
>   CPH1    CPH2
>
> In my top file generated with pdb2gmx I can see the improper dihedral
> entries for both
> NR1  CPH1  CPH2  H
> NR1  CPH2  CPH1  H
> Also after dumping the tpr file I can see IDIHS entries for both
> arrangement of atoms for the same molecule.
>
> According to Gromacs manual, the entry for improper dihedral i j k l is
> understood as follows
>          l
>          |
>          i
>        /   \
>      j      k
>
> So maybe I'm blind but I still don't really understand  the purpose of
> changing the middle atom types
> if there is in fact only one way these four atoms are connected in HSD. The
> only reason I see is to
> make it more rigid because I have "doubled" improper.

Again, I would not call it a "doubled" parameter, because they are 
clearly different in atom order. This usually implies a difference in 
stereochemistry; in this case, I don't know the history of why the 
impropers are assigned this way, but going back to the original CHARMM 
topologies there are in fact two impropers assigned around ND1, CD2, and 
CE1. The use of two terms with the same parameters but different 
connectivities is probably due to having to avoid some kind of 
asymmetry, though it seems unusual. Regardless, it is a faithful (and 
correct) representation of the force field and we have tested it for 
agreement between CHARMM and GROMACS.

-Justin

-- 
==================================================

Justin A. Lemkul, Ph.D.
Assistant Professor
Office: 301 Fralin Hall
Lab: 303 Engel Hall

Virginia Tech Department of Biochemistry
340 West Campus Dr.
Blacksburg, VA 24061

jalemkul at vt.edu | (540) 231-3129
http://www.thelemkullab.com

==================================================