Hi Ali! There is an option "-cut" where you can actually specify the distance to consider for molecules to be in a cluster. Otherwise you can even use g_cluster or g_dist. The latter one will calculate the distance between the centers of mass of two groups of atoms as a function of time, as it says in gromacs-manual.

Hi Inna Ermilova, thank you so much for replying to my question. I looked "Clustsize" and used "cut". This is my command:

gmx clustsize -f mdrunps.xtc -s mdrun.tpr -n Asph.ndx -hc histAsph_mol_1ns_.25.xvg -cut 0.25

This works perfectly when I want to compute the size distributions of atomic clusters. I wonder if you can help me with the definition of the "cut" command. Manual says the cut is the largest distance (nm) to be considered in a cluster.

Questions:

1. Does it mean the largest distance between any atom of my molecule? Is there any way that I can change that to the center of mass of my molecules?

2. When I add the "mol" option, it does not ask me the cluster between which molecules! I wonder if you have any idea?

By the way, I think g_distace does not work for me because I have 25 molecules that can aggregate and make clusters while g_distance measure distance between pairs of molecules. Also, I looked at g_cluster but I could not find any solution for my system. My main purpose is to calculate the attached formula. Any useful idea is appreciated

Dear Ali, I have used gromacs functions to calculate clusters and from the output it is clear that it takes into account any atoms/molecules. When you specify "-cut" the software "considers" the largest distance in a cluster. I guess, everything what is bigger than that won't be considered in a cluster. The default value is 0.35 nm. I think you shall measure the average size of your molecules before putting "-cut" value in your computation: it has to be reasonable.

Yes, 25 molecules is a big number for g_dist. If it is only one MD simulation I still would go for g_dist as it doesn't take too long and one can represent the data in a compact way. However, there is another function which can take into account several molecules simultaneously, it is g_mindist. By making an index file and specifying "-ng" you will get one file. You may then separate one file into 2-3 since you have 25 molecules.

Also for me the formula does not look too complicated, I think in the worst case you could try to code yourself the desired function.

Dear Inna , Thank you so much. I think g_mindist will works for me. I also trying some other methods. I will write here the result of my tests here in couple days.