[gmx-users] new residue went into crash during energy minimization

[Justin Lemkul]

On 7/7/16 10:08 PM, Wang Tao wrote:
> Hello,everyone!
>
> I am using charmm27 in gromacs for simulation of a double strand DNA , of which a residue is modified by a group.The forcefield parameters for the new residue is generated with swissparam.
> After conducting energy minimization , i checked the DNA's structure with pymol . Obviously , some thing is wrong .   Except for the atoms in the back bone , all the atoms in the modified residue are seperated from each other .
> What can i do to make it though the energy minimization?
>

"Separated" in a visualization program doesn't necessarily mean anything; 
sometimes programs guess bonded structure wrong.  The topology is definitive. 
If bonds are defined, they're present.  That's your first check.  Did you 
properly define the bonded connectivity in the .rtp entry you created?

I would also recommend (1) using CHARMM36 as it is improved relative to CHARMM27 
and (2) not relying on generalized force fields when trying to create integral 
residues in biomolecules.  Greater care needs to be taken to get a properly 
balanced representation of inter-residue interactions.  A modified nucleotide 
should be parametrized very specifically and very carefully.

-Justin

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

Justin A. Lemkul, Ph.D.
Ruth L. Kirschstein NRSA Postdoctoral Fellow

Department of Pharmaceutical Sciences
School of Pharmacy
Health Sciences Facility II, Room 629
University of Maryland, Baltimore
20 Penn St.
Baltimore, MD 21201

jalemkul at outerbanks.umaryland.edu | (410) 706-7441
http://mackerell.umaryland.edu/~jalemkul

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