GROMACS version: 2018.3
GROMACS modification: No
I am doing MD simulation of pure ethylene glycol system contating 512 molecules. I’m using charmm36 parameters obtained from MacKerell lab homepage. After EM using steep the PE curve converged to a positive value. I’ve found that same incindents were earlier reported. An explaination is the dominance of bonded interactions over non-bonded interactions. Is it OK to proceed with this or should I have to change or modify something? I’m in real dilemma and any kind of help is highly appreciated.
Thanks and regards,
Dhrubajyoti Maji
Hi,
you can proceed with MD equilibration and check the properties of your liquid model (e.i density) and compare with literature data.
kind regards
Alessandra
Thanks for your kind reply. I’ve already completed the eqlbn. After eqlbn I found the PE is still positive. The density deviation from reported simulated density is 1.92%. After production run other properties like diffusion coefficient, dihedral angle distribution etc show small but significant diffreneces from reported simulated values. Could you please shed light upon this matter? Thanks again.
With regards
Dhrubajyoti
Hi,
If you see differences in properties previously calculated using the same force field, then you can check your force field parameter implementation.
Otherwise, the difference are due to use of different force fields.
kind regards
Alessandra