Dear All,
We are working on parametrization of x non-standard (NR) residue using Charmm22, GLYCAM 06 (amber format) and GAFF force fields. Among these, we can able to parametrize them successfully in Charmm22 alone where we manually modified some of the DIHR angles to make it work. Indeed it works and we want to implement the same values in the respective dihedrals in GLYCAM as well as in GAFF but we see some major difference in the representation of GLYCAM and GAFF Dihedral types as follows compares to the Charmm22.
Ex:
In charmm22 we have changed the 0.0000 values to 109.40
@charmm22 dihedral types
; i j k l _ _ values
OG30 CG31 OBG5 CG31 1.0000 3 0.0000
OG30 CG31 OBG5 CG31 1.0000 3 109.40
and we want to implement the same 109.40 in GLYCAM and GAFF but we are not sure about which values to that we need to change among these " C0 ,C1,C2 ,C3, C4,C5", also don't know its a good idea to do the same.
@GLYCAM and GAFF
[ dihedraltypes ]
; treated as RBs in GROMACS to use combine multiple AMBER torsions per quartet
; i j k l func C0 C1 C2 C3 C4 C5
@glycam
O2 Cg Os Cg 3 8.53536 7.53120 11.54784 -16.06656 0.00000 0.00000
So if anybody knows how to convert these Charmm 22 DIHR to amber format kindly let us know your advice on the same would be really helpful.
Are you sure you want to do this? By setting the dihedral phase to a non-zero or non-180 value, you're making the parameter completely non-transferable. This is generally not done.
If you're sure that's appropriate (be very skeptical of it...) then use a tool like ParmEd to convert between formats.
Dear Justin Lemkul,
Thank you so much for the reply.
To answer your question yes.
Also we assessed those modified dihedral phases with modeller and it works. The reason we modified some DIHR phases because the initial non standard residue (NR) that we used for parametrization has linked with sugar in non 4C1 puckering conformation. So in order to have the x form of puckering conformations in our models we changed the DIHR and some IMPR phases. We are not sure it is correct but we can able to get the reasonable models with x form of puckering conformations along with 4C1 conformations.
Further we want to implement the same in simulation using GLYCAM and GAFF. So sure I will try ParamEd.
Thanks again.
Mathematically, you can set whatever you like in terms of phase. Strictly speaking from the force field conventions, it's not correct. If your conformation deviates, you'll get unpredictable results that may not have physical meaning.
- Badges
- Science topic