Gmx hbond output with -ac and -life options

User discussions
1

GROMACS version: 2020
GROMACS modification: No

Hello,

I am using gmx hbond with the -ac and -life options to determine the average hydrogen bond lifetimes in my system. When I run the command, I get the following output:

Type Rate (1/ps) Time (ps) DG (kJ/mol) Chi^2
Forward 0.169 5.908 8.931 1.48156
Backward 0.000 6843.424 26.420
One-way 0.332 3.009 7.259
Integral 0.603 1.657 5.780
Relaxation 5.025 0.199 0.526
HB lifetime = 1.03 ps

From my understanding of the GROMACS documentation and the provided literature, the lifetime can be calculated as the inverse of the forward rate constant. However there is a discrepancy between the value obtained in this way (5.908 ps) and the one provided in the last line of the output (HB lifetime = 1.03 ps).

I have also tried, as indicated in the documentation, doing a multiexponent fit to the autocorrelation function output and integrating the obtained function to get the average hydrogen bond lifetime. The value I get is also very different from both values obtained from the output.

Accordingly, I would really appreciate it if someone could help answer the following questions:

  1. What is the reason for the discrepancy between the two values (5.908 vs 1.03 ps) obtained from the GROMACS output?

  2. What are the one-way and relaxation values obtained in the output?

  3. Is the lifetime value obtained using the multiexponent fit + integration approach more meaningful than those obtained automatically from the output?

  4. What considerations need to be taken into account when using the gmx hbond utility to analyze hydrogen bonding between two protein chains?

Thanks in advance!