[gmx-users] Details of g_rdf functions in version 4.6.5.

[Henry Vider]

From: Henry Vider
Sent: Thursday, June 13, 2019 11:37
To: gromacs.org_gmx-users at maillist.sys.kth.se
Subject: Details of g_rdf functions in version 4.6.5.

Dear GROMACS mailing list,

I am writing here due to the inconsistencies in my Radial Distribution Functions (RDF) calculations. Currently I am using v. 4.6.5. of GROMACS.

I am simulating water-organic binary solvents with a few solute molecules added into the system. I have simulated a lot of different compositions for water and organic solvent, but I have kept solute molecule number constant at 10 molecules. For example, one of the systems is 10 Ethyl Acetate, 1250 Ethanol and 1240 Water, totaling 2500 molecules. (As a side question, is this system size appropriate for studying solute-solvent interactions when I have 10 solute molecules or should I lower the number of solute molecules?)

When initiating g_rdf function with default settings, i.e. only specifying -f *.xtc , -s *.tpr and -o *.xvg and -cn, I am prompted with a selection for two groups. In the case of example system, I choose firstly the Ethyl Acetate group and then Ethanol group which means I should get RDF of Ethanol molecules distributed around Ethyl Acetate molecules. However, when looking at the graphical representation of the RDF, I notice that the function starts at 0.2 nm which is unreasonably close for molecule. I have come to a conclusion with the help of GROMACS manual and looking at the short description of parameters for the g_rdf function displayed when initiating in linux terminal, that if I do not specify -rdf with something, for example mol_com etc, the RDF-s are calculated with atoms.

My question is, what atoms is the script choosing automatically if I do not specify -rdf value? I tried looking at gmx rdf script, but I could not understand.

Secondly, in order to get RDF-s calculated for COM Ethyl Acetate and COM Ethanol, what should be the command line for g_rdf function? I have tried several things but I am getting different results, sometimes even RDF-s are abnormal but CN functions seem appropriate.

Thank you in advance!

Faithfully yours,
Henry