From: Dalhaimer, Paul Matthew (pdalhaim_at_utk.edu)
Date: Sun Feb 20 2022 - 19:03:16 CST
Excellent information. Thank you.
From: Giacomo Fiorin <giacomo.fiorin_at_gmail.com>
If one wanted a continuous description of the chemical reaction, then it wouldn't be just a matter of changing the source code, but rather defining a suitable force field to describe the reaction. Biomolecular force fields like CHARMM or AMBER are not designed to simulate reactions. Of those that do, I don't think there are any that provide the same combination of low computational cost and decent accuracy of the intermolecular forces that are most important in biophysics. Even DFT (i.e. including the electrons to a certain degree) kind of stinks at describing intermolecular forces, unless you use a very modern functional that is also more expensive and difficult to use in a transferable manner.
So basically, if one accepts that there isn't a good enough simulation model to describe the thermodynamic cost of a continuous reaction, simulating the two endpoints separately (with and without the disulfide bridge) with standard force fields is what is usually done.
Giacomo
On Sun, Feb 20, 2022 at 9:08 AM Dalhaimer, Paul Matthew <pdalhaim_at_utk.edu<mailto:pdalhaim_at_utk.edu>> wrote:
This archive was generated by hypermail 2.1.6
: Tue Dec 13 2022 - 14:32:44 CST
Date: Sunday, February 20, 2022 at 7:37 PM
To: NAMD list <namd-l_at_ks.uiuc.edu>, Dalhaimer, Paul Matthew <pdalhaim_at_utk.edu>
Subject: Re: namd-l: S-S bond
Hi Paul, you can manually add an empirical bond via extraBonds:
https://www.ks.uiuc.edu/Research/namd/2.14/ug/node29.html#SECTION00086400000000000000
Although you can't do it during a simulation, you can certainly read binary restart files, turn on the new bond, and continue from there. What isn't obvious at all is how to deal with the hydrogen atoms, or to update the charges and vdW coefficients of the atoms involved in the reaction to reflect the changing chemical environment. One is probably better off switching between topologies, as is done for example in the NAMD implementation of constant pH simulations.
All,
Has anyone tried to modify the NAMD source code to allow for covalent bond formation during a simulation? Cysteine-cysteine, would be an obvious target to try this on.
-Paul