I'm trying to add a molecule in aminoacids.rtp (new user)
By looking at the pre defined amino acid entries and the chapter 5 of the manual I started building it up. The fields that the file include for the [ atoms ] field are: 1. atom type, 2. "code name" taken from atomtypes.atp, 3. charge, 4. charge group.
My first question is: I know that in theory charge group is grouping atoms in order to get zero or an integer charge and that it has to be with as few atoms as possible. But aren't there any additional criteria for grouping atoms? I don't think that I have to make an arbitrary choice just for keeping an integer type of charge.
The second and most important question: the [ bonds ], [angles], [dihedrals], [improper] fields in the existing entries of aminoacids.rtp are defined only by atom type. I don't think I can use this format because an atom type is located in different location in my molecule, can be bonded with different atoms and has different charges. So I believe I have to use a type of numeration to define the proper bonding. With that in mind which is the proper syntax to do this in a proper way?
P.S.: I know that my questions are a bit obscure and hard to follow, sorry about that, not a native English speaker. Please feel free to ask me for any additional info necessary. Thanks in advance
Let's clarify the contents of the [atoms] directive of the .rtp. The fields are:
1. atom name
2. atom type
3. charge
4. charge group
The charge groups should be small and chemically intuitive. You wouldn't hack together atoms from different parts of a molecule to fit some preconceived notion; the groups correspond to actual functional groups. Note, though, that if you are using a current version of GROMACS with the Verlet scheme, charge groups are totally irrelevant. Most force fields do not use charge groups, anyway.
The [bonds] section is defined by atom name. Names are always unique in a residue definition (they have to be) so this is actually very straightforward to do.
Thank you very much. I must have seen such files from previous versions of Gromacs.
If it is not much to ask, could you post a "dummy" scheme of how to define bonds?
Do you by any chance know a good workshop - tutorial on how to perform these actions? The ones I found contained the "unnecessary" info I posted above.
See Chapter 5 of the manual. It is the authoritative source for this information. See also existing .rtp entries for a guide. Defining bonds really is as easy as:
[ bonds ]
A B
B C
...
- Badges
- Science topic
- Similar topics
- Physical Sciences
- Atomic, Molecular and Optical Physics
- Atoms