I would like to calculate Hydrogen Bonding (HB) interaction from simulation a simulation.
Let's say I have two molecules (molecule A and molecule B) with several OH groups in each molecule.
At first, I make molecule A, which contains heavy atoms such as oxygen, to be the "acceptor" and molecule B to be the "donor". Then, I make the molecule B ( also have several OH groups) to be the "acceptor" and molecule A as the "donor".
In both condition I calculate the HB and I have two values of HBs. For getting the average HB for the system, is that possible to add both the values or calculate the average from both the values?
Appreciate any explanation to this. Thank you.
GROMACS provides a command to accomplish such task:
http://www.gromacs.org/Documentation/Gromacs_Utilities/g_hbond
Yea thanks. But I want to know the logic behind the HB calculation.
http://en.wikipedia.org/wiki/Hydrogen_bond
It is based on time averaged monitoring of both length and angles between the acceptor and the donor atoms. It should be easy to calculate once atomic coordinates became avalaible.
May be I missed some info in my question. I am using AMBER. There I use ptraj or cpptraj to do the calculation. In there, if I define molecule A to be donor and molecule B as the acceptor, at the end of the calculation I get a value. On the other hand, if I define the donor and acceptor vice-verse, I also get some value which is not the same with previous one. So here I get the confusion, do I need to add them together or get the average of them.
You do not average them. Even if you deal with non-bonded interaction between A and B, in one case A is an acceptor and in the other case A acts as donor. These cases describe two different kind of Hbond.
Therefore, the residue pair A:B is connected by two hydrogen bonds that switch each other during the simulation. Perhaps, for geometrical reasons, the former will appear more stable than the latter.
Besides nice replies you already have, I suggest you to have a look to NCIPlot, a very efficient/easy/free code that provides a qualitative analysis of any noncovalent interaction in complex systems and, particularly, where H-bond networks are presented.
Ref: J Chem Theory Comput. Mar 8, 2011; 7(3): 625–632
Hope it helps. Best,
Jose
I'm only familiar with electronic structure simulations. But, have you tried an statistical mechanics approach? I mean simple things like weight every HB by its probability density in a Boltzmann distribution.
A function that "weights" the strength of the Hydrogen Bonding can be defined.
There are many different geometrical definitions, most of them based upon X...H distance or some angle.
Here there's an application of one of such functions (F_HB):
http://pubs.acs.org/doi/abs/10.1021/ct100625e?prevSearch=%255BContrib%253A%2Bmuniz-miranda%255D&searchHistoryKey=
Hi
could you calculate hydrogen bonding from simulation?
did you use Molecular Dynamic (MD) simulation?
- Badges
- Science topic