Improper Dihedrals in HIPPO

Discussions about the force fields implemented in Hippo. For general force field discussions see Molecular Simulations -> Force fields.
Post Reply
ldav
Posts: 8
Joined: Wed Jul 23, 2008 9:38 am

Improper Dihedrals in HIPPO

Post by ldav »

Regarding DIHEDRALS/IMPROPER DIHEDRALS. I am not sure how improper are taken into account by Hippo. Is there an easy way to go from the harmonic formulae K(theta-theta0)**2 to the V1-V3 coefficients?
jakob
Developer
Posts: 6
Joined: Sun Jul 06, 2008 4:54 pm

Re: Improper Dihedrals in HIPPO

Post by jakob »

The dihedrals are treated exactly in the same way as BOSS/MCPRO, including the impropers.

Use a V2 torsion only if planarity of geometry is desired, as example, look for e.g. ester torsions, or ring torsions. This approach of course only deals with problems involving theta0 = pi (planarity)


There is an arbitrary dihedral constraint energy that can be used in MC simulations (MD not yet). This will put an harmonic constraint on that dihedral, with a chosen theta0. At the moment, it is only implemented for enforcing helicity in phi/psi angle of a protein residue.

You can switch this on in the input PDB file, format is
REMARK hippo: restrain helical 5X1X4X - 5X1X4X k0

for example
REMARK hippo: restrain helical ALA A 23 - CYSH A 287 40.0

will set phi/psi backbone torsions of residue ala23 chain A to res cysh287 chain A. Restraining is for ideal a-helix, and the force constant here is 40 kcal/mol*rad

you can also use several lines, to restrain individual helices.
The energy files have an entry E_rst for the restraint energies.
Note that the restraints do not enforce immediate helicity, sampling issues prevent this from happening.

The following is taken from a VPU test run. Have a look at the log.txt file to see if the correct backbone torsions are constrained.
REMARK
REMARK hippo: restrain helical MET 2 - ARG 31 40.0
REMARK

In case you really need it, I can program a different loading syntax that allows individual dihedrals to be constrained to arbitrary theta0. I havent used that yet, because I always wanted to constrain only helices.

hope this helps,
Jakob
Post Reply