I am trying to simulate a transition state in explicit solvents using GROMACS 4.6.5. GAFF parameters have been generated with the help of Gaussian 09 and AMBER tools. After NVT equilibrium the transition state structure is breaking apart. How can I keep the integrity of the transition state structure through out the simulation?
Mr. Deb,
Could you specify which parameters you have generated by Gaussian 09?
Geometry optimization and RESP calculations have been done using Gaussian 09. Finally GAFF parameters have been generated using the Gaussian out put file in AMBER tools.
What do you mean by breaking apart? There cannot be any breaking of bonds in MD. Did you check for pbc issues?
Hi Dr. Dantu, in transition state structure, the bonds between the reactants (formation/breaking of these bonds led to product) are not real bonds. Distances between the atoms involved in transition state are longer than common C-C bond (in my case).To do simulations we need to do some thing such that the relative orientation of the reactants remain same through out the simulation as it was in the initial transition state structure.
After doing NVT, the two reactants are going far apart since there is no real bond defined in topology of the transition state structure. I need to keep the integrity of transition state complex so that the whole TS can move but the reactants won't go apart from each other (i.e. distances, angle, etc of the TS structure would remain same).
You can try distance restraints, with broad well potentials. They will keep the distances intact. You can also define dihedral restaints I suppose. Check the gromacs manual.
Define the force field parameters (atomic types, atomic charges, bond lengths, bond angles, and torsion angles) of the entire transition state structure, i.e., including the force field parameters of all "partial" bonds.
Hi Martin,
I have defined atomic types, atomic charges, bond lengths, bond angles and torsion angles . But in bond angle and torsion angle section, other parameter are present (cth, c0, c1, c2, c3 , c4 & c5). How can I get the values for these parameter for the atoms which are involved in transition state? Do I need to form a separate itp file for the atoms involved in transition state?
Mr. Deb,
When you have used Gaussian, why you do not employ, for examples, AMBER-IRC for TS computations, among others, by the way?
You could pay attention to the attachment (Amber.gif). It is given an example of AMBER-IRC of a complex {[Zn(C12H26O6)]2+}.
The route sections are:
1)
#p opt=calcfc freq amber test
2)
#p amber irc=(cartesian,rcfc) iop(1/45=1) geom=check guess=read test
For IRC computations you could read the IOp manual of Gaussian (Section Iop(1/45)). The manual is attached as well.
Thanks a lot Prof. Ivanova. I will perform the Gaussian calculations once again.
- Badges
- Science topic