AMBER Archive (2005)

Subject: Re: AMBER: about the %occupied of hydrogen bond

From: Carlos Simmerling (carlos.simmerling_at_stonybrook.edu)
Date: Mon Sep 19 2005 - 08:04:14 CDT


that's what I thought, so look at my original suggestion (Hbond vs time for
each). you'll need to postprocess the data or find a different program.
in the particular case that you mention do you really need to know the
total % if the ID1 is already 98.4%? That is almost all of the time.
Unless you are very sure that your data is well converged and independent
of initial conditions, you probably have uncertainties larger than +/-1.6%
anyway.

haixiao jin wrote:

> thank youfor your help! i should make my question more clear.
> i calculated the H-bond occupied by PTRAJ in AMBER. the cutoff of
> distance and angle of hydrogen bond is 3.5 and 120.0, respectively.
> i know the mean of "%occuped" between two atoms, for example, the
> %occupied between the O6 of ligand and the OD1 of ASP184 is 98.4%. it
> is presents the percentual fraction of existence of the hydrogen bond
> out of the total time of simultaion is 98.4. but there also exist the
> hydrogen bond between the O6 and the OD2 of ASP184. so, how can we
> express the total %occupied of H-bond between O6 and ASP184?
>
>
>
> On 9/16/05, *Carlos Simmerling* <carlos.simmerling_at_stonybrook.edu
> <mailto:carlos.simmerling_at_stonybrook.edu>> wrote:
>
> maybe the original question is how to determine the % of time
> these two
> GROUPS
> are hydrogen bonded, and ptraj gives the % for the specific atom
> pairs.
>
> For example, if the two atom pairs are H-bonded 50% each, it could
> be that
> 50% of the time there is a bifurcated Hbond and 50% no Hbond, or
> it could be that It is 1 Hbond in 50% and the other in 50%, so 100%
> of the time there is an Hbond between the groups. There just isn't
> a way
> to tell that from the average data, you'll need to do some
> processing of
> the
> Hbonds vs time and see how often you have at least 1 of them.
>
> If this isn't what you meant, please try to clarify exactly what you
> want to calculate.
>
>
> Thomas E. Cheatham, III wrote:
>
> >>I have gotten the hydrogen bond occupied. But in the followed
> two example, I
> >>don't know how to express the %occupied.
> >>
> >>atom
> >>
> >>residue
> >>
> >>distance
> >>
> >>%occupied
> >>
> >>WAT471-O(H1)
> >>
> >>ASP220 (OD2)
> >>
> >>2.697
> >>
> >>55.20
> >>
> >>
> >
> >>From the information given, it is not clear where you obtained
> the data
> >shown, but if this was the hydrogen bond facility of ptraj, you
> in the
> >above observed an average distance (when formed, i.e. < cutoff)
> of 2.697
> >angstroms and an occupancy of 55.2% (meaning that it was formed
> slightly
> >more than half the time). Occupancies and average distances will
> depend
> >on the definition of the h-bond (cutoff distance, angle). If the
> distance
> >cutoff is longer, the occupancy may be higher.
> >
> >
> >-----------------------------------------------------------------------
>
> >The AMBER Mail Reflector
> >To post, send mail to amber_at_scripps.edu <mailto:amber_at_scripps.edu>
> >To unsubscribe, send "unsubscribe amber" to majordomo_at_scripps.edu
> <mailto:majordomo_at_scripps.edu>
> >
> >
>
> -----------------------------------------------------------------------
> The AMBER Mail Reflector
> To post, send mail to amber_at_scripps.edu <mailto:amber_at_scripps.edu>
> To unsubscribe, send "unsubscribe amber" to majordomo_at_scripps.edu
> <mailto:majordomo_at_scripps.edu>
>
>

-----------------------------------------------------------------------
The AMBER Mail Reflector
To post, send mail to amber_at_scripps.edu
To unsubscribe, send "unsubscribe amber" to majordomo_at_scripps.edu