[gmx-users] Re: g_analyze

[Justin Lemkul]

On 11/10/13 7:18 PM, bharat gupta wrote:
> I checked the file hbnum.xvg file and it contains three columns - time,
> hbonds, hbonds that donot follow the angle criteria. In that case SS1 is

The third column is not actually H-bonds, then ;)

> the average of actual hbonds (2nd column ) and SS2 is the average of 3rd
> column. Am I right here or not ??
>

Yes.

> I tried to calculate the h-bond for residues 115-118 individually, and then
> checked the average for each residue. For single residue calculation, the
> g_analyze average value is correct.
>
> But when I calculate the h-bond as a range 115-118, I get the g_analyze
> value as 1.62 . I calculated the average manually in excel, got the average
> values as 16.2 [which is (g_analyze avg value)/10].
>

That is impossible.  You cannot get a different average by examining the same 
numbers.  Read the g_analyze output again - I am willing to bet that you're not 
seeing the exponent of the scientific notation.

> I then added up the average values of h-bonds of individual residues and
> the final comes around 16.2, same as that of the 115-118 range h-bonds.
> This means that my calculation is correct.
>
> I also used trjorder to calculate h-bond at distance 0.34 for residues
> 115-118. I got the average value around 2.51 from g_analyze, where as
> manual calculation gives 25.1. I don't knw why for the range the g_analyze
> give avg as (actual avg value)/10 ??
>
> Why does trjorder and g_hbond gives different number of hydrogen bonds for
> the same residue set??
>

All of this comes down to correctly reading the screen output.  I have no idea 
what you're doing with trjorder, though.  It doesn't measure H-bonds.

-Justin

-- 
==================================================

Justin A. Lemkul, Ph.D.
Postdoctoral Fellow

Department of Pharmaceutical Sciences
School of Pharmacy
Health Sciences Facility II, Room 601
University of Maryland, Baltimore
20 Penn St.
Baltimore, MD 21201

jalemkul at outerbanks.umaryland.edu | (410) 706-7441

==================================================