Date: Thu, 03 Apr 2003 14:46:52 -0800
From: Holger Gohlke
Subject: Re: Side-chain/main-chain decomposition in MM-GBSA

> Dear AMBER commmunity,
> I'd like to ask about the way the decomposition to side-chain and main-chain
> contributions is done in MM-GBSA calculations.
>
> I assume for protein (let's say in parm99) it's straightforward; N,H,C,O
> atom types are considered main chain, the rest is side chain.

In case of protein residues, N, H, C, O, CA, and HA (and HA2,3) are
considered backbone, everything else sidechain. Nucleic acids are not
yet decomposed.

> But how is it for ligands and gaff? Is it n,hn,c,o? Can it tailored (so as
> to add some atom types)?

Ligands are not yet decomposed - they are considered to be "sidechains".
Currently, what is considered to be backbone is hard-wired in the DATA
IGRAP and DATA ISYMB statements at the beginning of the subroutine
RGROUP in rgroup.f . For the moment, by extending these definitions, it
should be possible to adapt the decomposition scheme.

Best regards

Holger

>
> Or is prepin Main chain information taken into account?
>
> Thank you in advance for your responses.
>
> All the best,
>
> Martin