[gmx-users] g_hbond errors

[YOLANDA SMALL]

Hello All,

I am having some problems with getting g_hbond to give the correct number of
h-bonds.  Using the -sel option with an index file containing triples crashes
with a segmentation fault.  To get around this, I used the -n option with the
same index file containing triples and it processes the results but they seem
to be using an incorrect geometric criteria for determing h-bonds. 

I am comparing the results to the h-bonds that I can visualize in VMD for the
same trajectory.  If I decrease the angle criteria for g_hbond I am able to get
close to the same number of h-bonds that I see with VMD; however, this seems
like a glitch of some sort.  I am using the following command:

g_hbond -f file.trr -s file.tpr -n select.ndx -num -ac -a 35 -r 0.33 -nomerge
-hbn

My selection index file looks like this:
[ SEL ]
1838 1839 3595 

Can anyone shed some light on what may be going wrong??? 

Thanks,
Yolanda