Hi,

have you checked the -selrpos and -seltype? Probably with -seltype res_com (or whole_mol_com) you will get what you want.

The -selrpos is the reference position, the default is 'atom', so if you do not select an atom for your reference position (e.g. you select a group of atom), you need to specify which type of center you want to, and which part of the molecule (residue, part of the molecule, whole molecule) is it.

The -seltype is the type of selection for the atom or group, residue, etc, you want to calculate the rdf for, in relation to your reference position.

Other option is to use -sel, as a selection to calculate RDFs, in relation to your reference position. As when you select in vmd graphic selection (remember that gromacs use nm and vmd angström, as default units).

Thank you very much for your time and answers.

Unfortunatelly I still can't solve the issue. I calculaten the function I want to receive on the oter computer with older version of GROMACS and now i'm trying just to solve it.

My system:

1) Whole system

2) Dendrimer

3) Counter-ions

4) Water

Please correct me if i'm wrong.

1) If I want to calculate the rdf of an atoms (densities) with respect to the dendrimer center of mass then I should cpecify it with an obvious flag "-selrpos whole_mol_com", right? If I start it like this, then it takes enormous amount of time (more than 1000 times slower than with "-com" flag in previous GROMACS version). It seems to calculate the pair distribution function, which, of course, takes much longer time.

2) Other option is to set the second flag "-seltype" in addition to "-selrpos". And here no matter what I choose (exept atom, of couse) I have just the straigh line as output. But here the time of data processing seems to be correct.

And the second pitfall is the choice after you push the enter. I have to do two selections - for "ref" and for "sel". In both I choose the Dendrimer because I want to have the dendrimer density (the rdf) with respect to it's mass center. I'm in doubds I making mistake here, but... I'm already not sure of anything.

---------Small Note---------

It is also interesting that it does the same with this two command lines:

1) "gmx rdf -f *center.xtc -s trp_nw.tpr -o density.xvg"

2) "gmx rdf -f *center.xtc -s trp_nw.tpr -o density.xvg -selrpos whole_mol_com"

3) "gmx rdf -f *center.xtc -s trp_nw.tpr -o density.xvg -selrpos whole_mol_co"

4) "gmx rdf -f *center.xtc -s trp_nw.tpr -o density.xvg -selrpos whole_mol_c"

etc.

With no warnings and errors

------------

I will be very thankful for any answers and thank you for reading such long text.

A strong coffee to all of you ;)

-Boris.

You might try:

gmx rdf -f *center.xtc -s trp_nw.tpr -n index.ndx -tu ns -ref 'resname "(dendrimer name)"' -sel 'resname "(your selection name in the index file)"' -pbc -o outputfile

Your selection name in the index file may also be the dendrimer. I think that it could work.

Thank you for the coffee, it is ok for me ;)

Thank you for the answers!

I decided to install the 5.0.7 version and now i'm working with "-com" flag and everything is ok.

The 2018 GROMACS seems to have bugs =)

M. Florencia Martini can I cross check something with you, please.

My system consists of a protein embedded in a membrane with explicit water and ion environment. I would like to compute the distribution of the COM of lipid molecules to the surface of the protein.

In my first test I used the following command:

gmx rdf -f traj.xtc -s md.tpr -n index.ndx -o rdf_T1.xvg -surf mol -seltype whole_res_com

This command generated an error:

Inconsistency in user input:

-surf only works with -ref that consists of atoms

When I run again the same command line but with -selrpos, it works "fine"

gmx rdf -f traj.xtc -s md.tpr -n index.ndx -o rdf_T1.xvg -surf mol -selrpos whole_res_com

I know that COMs are not supported with -surf. But -seltype is supposed to be related to "the type of selection for the atom or group, residue, etc, we want to calculate the rdf for". I can't understand why I got this error when using -seltype with -surf option. Any comment on this?

In another test I tried this:

gmx rdf -f traj.xtc -s md.tpr -surf mol -o rdf_g15.xvg -ref 'protein' -sel 'com of group 15'

this one works fine, faster but the output is different from the previous command.

but I am not sure if I am calculating the right thing.

Does the command lines that I am using are producing the RDF of COM of lipids to the surface of the protein?

I would really appreciate your help and opinion.

Thanks a lot.

The last command line (gmx rdf -f traj.xtc -s md.tpr -surf mol -o rdf_g15.xvg -ref 'protein' -sel 'com of group 15') measures the radial distribution function of the center of mass of the lipid group (all together), relative to the surface of the protein. You need to use (gmx rdf -f traj.xtc -s md.tpr -surf mol -o rdf_g15.xvg -ref 'protein' -sel 'mol_com of group 15'), in order to consider the mean of the COM of each lipid molecule.

Cheers

M. Florencia Martini Thanks a lot for your reactivity and help!

All the best for you :))