Dear colleagues researchers,
Could you help me with a problem in MD analysis using Gromacs?
I'm using Gromacs version 2021.4 now, and trying to analyze hydrogen bonds (using HbMap2Grace) with the command:
gmx hbond -f ../md_center.xtc -s ../md.tpr -n ../index.ndx -num hbond-lig -r -0.4 -hbm hbmap-lig -g hb-lig -life hblife -tu us
The program gives me the following error:
Fatal error:
Not enough memory. Failed to calloc -79 elements of size 4 for rdist
(called from file /home/priscila/Documents/gromacs-2021.4/src/gromacs/gmx_hbond.cpp,
line 2856)
I've also tried it with version 5.1, 2018, 2019, 2019.6 and 2020, but all with the same problem.
P.S.: I don't think it's a lack of physical memory on my computer, cause there is enough storage space.
Dear Priscila Goes Camargo ,
Some options of hbond tool requires a lot of RAM.
At the end of the "gmx hbond -h" command:
"Note: options -ac, -life, -hbn and -hbm require an amount of memory proportional to the total numbers of donors times the total number of acceptors in the selected group(s)."
You can try to analyze a smaller number of frames or restrict the selection of donors/acceptors.
Best,
Felippe
Dear Priscila Goes Camargo ,
1- Options involving autocorrelation require a lot of memory and it is not recommend for long trajectories, it is because the lifetime of the H bonds are short in relation to simulation time of typical trajectory of a large system;
2- Gromacs 200.6 is recommended for building maps. I'm not sure, but many surface problems have been fixed from this version.
Best,
Marcello
Dear colleagues, I realized that sign - (minus sign) in front from 0.4 value it's the problem in my work.
When I used the command:
gmx hbond -f ../md_center.xtc -s ../md.tpr -n ../index.ndx -num hbond-lig -r 0.4 -hbm hbmap-lig -g hb-lig -life hblife -tu us
It works.
Thank you for your recommendations.
- Badges
- Science topic