Water forming more than 4 bonds

User discussions
1

Hi all,
I want to calculate the number of hydrogen bonds formed by a single water molecule with my cosolvent. For the same, I have used gmx hbond and analysed the index file generated (gmx hbond -f *.xtc -s *.tpr -hbn *.ndx).
Under the [hbonds_cosolvent-SOL] section, I could see atom no 21833 (corresponding to oxygen of water molecule) being mentioned 5 times:
90 93 21833
149 147 21833
21833 21834 79
21833 21834 80
21833 21834 173
By which, I inferred that water is forming more than five hydrogen bonds, of which 3 bonds are formed by hydrogen of oxygen. This seems to contradict with basic chemistry. Could you please suggest how we can interpret the results or am I missing something?

2

HI,
I guess you refer to one of the output file of gmx hbond.
The tools output all the observed hydrogen bonds, but this does not mean that the hydrogen bonds occur at the moment in time.

To verify which hbond occur at the same moment in time, you can plot specific hbond vs time.

Best regards
Alessandra

3

Thank you for your reply. I would like to mention that I have used -b and -e flags for the same time point. That’s why I am not being able to understand the occurrence of 5 bonds.
Looking forward to your suggestions.

Thanks
Nidhi

4

The output hbond.ndx is best used in conjunction with hbmap.xpm. The fact that 5 H-bonds occur during that time frame does not mean they occur simultaneously. It just means that each exists at some point during the specified analysis timeframe. Plot hbmap.xpm and you will see at what time frames each H-bond occurred.

5

Thank you Justin for your reply. When I analyzed hbond.ndx and hbmap.xpm in conjunction, I could see that only three hydrogen bonds were formed at t1 (=268600ps) time point. The result was consistent with VMD analysis as well. But could you please help me understand why the other two hydrogen bonds appeared in the hbond.ndx file since in the command line (gmx hbond -f *.xtc -s *.tpr -b 268600 -e 268600 -hbn -hbm -hbnum) I have mentioned my specified timeframe as just one time point (=268600ps)? What is meant by some point during the specified timeframe?

Thanks
Nidhi

6

To me more specific, I guess the problem lies in creating an index file for performing hydrogen bond analysis.
When I use
gmx hbond -f *.xtc -s *.tpr -hbm -hbn -hbnum -b 268600 -e 268600
and select water and cosolvent as two groups between which I would like to compute hydrogen bonds, I could see atoms of a single water molecule appearing five times in the hbn.ndx (consistent with hbm.xpm file when analyzed in conjunction).

90 93 21833
149 147 21833
21833 21834 79
21833 21834 80
21833 21834 173

But when I analyzed those water molecules separately by creating index file for each, eg,
a90|a93
a21833
and select these two groups for hydrogen bond analysis and subsequently all other pairs one by one, then only three hydrogen bonds appeared
(a21833|a21834 and a79
a21833|a21834 and a 80
showed no hydrogen bonds)
My question is: Is this a bug in the software or do we have to make index groups separately (second procedure of creating index file, and we can’t select water and cosolvent as index groups for analysis)?

Thanks
Nidhi

7

Instead of this approach, can you repeat the analysis on a single coordinate file from gmx trjconv -dump 268600 rather than with gmx hbond -b -e?

Highly unlikely. This is normal usage of gmx hbond.