News:

SMF - Just Installed!

Main Menu

about the Torsion in molecule.def file

Started by raspa666, June 12, 2023, 09:14:10 AM

Previous topic - Next topic

raspa666

For example, in the 2-methylbutane.def, there is a section of '# Torsion n1-n2-n3-n4 type'

# Torsion n1-n2-n3-n4 type
0 1 2 3 TRAPPE_DIHEDRAL   -251.06  428.73  -111.85  441.27
4 1 2 3 TRAPPE_DIHEDRAL   -251.06  428.73  -111.85  441.27

How can we get the 4 value of  TRAPPE_DIHEDRAL(-251.06  428.73  -111.85  441.27)?
Is there a software that can output a data file containing this value?

Thank U! :)

plgmattpokora

Hi,

the parameters that describe the torsion angles (as well as other parameters describing both the bonding and non-bonding interactions) are a part of a specific forcefield, in the example you provided, the TraPPE forcefield. You did not specify what molecule you are interested in. You can find the TraPPE forcefield description here: http://trappe.oit.umn.edu/. You can find there, for example, 2,3-dimethylbutane, which shares the dihedral parameters with 2-methylbutane.

Bear in mind, however, that description of a molecule within a forcefield is not an easy task. You can also try the OPLS-AA forcefield in a case you cannot find your molecule within the TraPPE, here is a generator of the parameters: http://zarbi.chem.yale.edu/ligpargen/

In either case please remember to make sure you use the proper units of the parameters in RASPA. It may be also a good thing if you read a bit about the forcefields in atomistic simulations.

Cheers,
Mateusz

raspa666

Thank U, Mateusz´╝ü :)

I am interesting in a moluecule --- CH3COOH, however, i can not find a mol.def file in the raspa-molecules. So I try to write a def file.

emmm, you mentioned that 'use the proper units of the parameters ', but if I try the OPLS-AA forcefield, how can I tranfersform the parameters/unit to a right one?

I have no idea...

Hoping you response.

Thank U ! :)

plgmattpokora

#3
Hi,

you have to look for the units in the RASPA2 manual. For example:

HARMONIC_BOND

describes bonds and needs two parameters with units K/A^2, A.

From the LigParGen server, you can get the topologies in different formats. I used the GROMACS itp file, the units used in GROMACS can be found here: https://manual.gromacs.org/current/reference-manual/topologies/topology-file-formats.html

So GROMACS uses units kJ/(mol*nm^2), nm. The conversion from nm to A is straightforward, but in the first case we need:
a) convert from kJ/mol to J (multiply by 1000 and divide by Avogadro's constant),
b) convert from nm^-2 to A^-2 (divide by 100),
c) convert from J/A^2 to K/A^2 (divide by Boltzmann's constant)

that brings us to multiply the value from GROMACS' itp file by 1.2037. In the same manner, you would obtain SIX_COSINE_DIHEDRAL potential in RASPA2 from GROMACS' RB dihedral parameters values by multiplying them by 120.272355. Please refer to the manual of RASPA2. And also please be aware that if you have no experience with such simulations, it can take a few days (or even weeks) to understand how to prepare inputs properly and run the simulations correctly.

Also, as you will be preparing the forcefield description, please pay attention to the Lennard-Jones parameters. You need to perform such a conversion as well. In some cases, like in the case of the Dreiding forcefield, you also need to check if the values are scaled in relation to the equation used in RASPA2. You need to read the manual and papers to understand everything correctly, and that can be challenging.

Cheers,
Mateusz

raspa666

Wow! Thank you very much, Mateusz!

It is really helpful !!! :) :) :)

I know how to deal with the data now. :)