How to fix RMSD analysis results?

User discussions
1

GROMACS version: 2019
GROMACS modification: Yes/No
Here post your question

I’m analyzing RMSD results for free energy perturbation (FEP). I execute a command below, corresponding to 2 loops but the graph results do not match each other (chart below), how can I fix it?

gmx trjconv -f md_0.xtc -s md_0.tpr -ur compact -pbc mol -center -o traj_0.xtc -n Pro_MOL_noH.ndx
gmx rms -f md_0.xtc -s md_0.tpr -o rmsd_noH_0.xvg -tu ns -n Pro_MOL_noH.ndx

dsfd
dsfd1787×940 43.9 KB

2

Based on the results you have provided, the initial guess might be the PBC issue. I recommend visually inspecting your system using VMD to get a better understanding. If it is PBC issue you can fix it by GROMACS 3-step workflow.

3

I have changed the command like:

gmx trjconv -f md_0.xtc -s md_0.tpr -ur compact -pbc nojump -center -o traj_0.xtc -n Pro_MOL_noH.ndx
gmx rms -f md_0.xtc -s md_0.tpr -o rmsd_noH_0.xvg -tu ns -n Pro_MOL_noH.ndx

yet, i still got two different graph for RMSD.

Please correct me if i am wrong.

4

Your xtc file may be corrupted. You may have to rerun the simulation

5

It is almost certainly a PBC issue and not a problem with the xtc file.

With your gmx trjconv command you have generated a new trajectory, called traj_0.xtc. You need to use that as input to the gmx rms command:
gmx rms -f traj_0.xtc -s md_0.tpr -o rmsd_noH_0.xvg -tu ns -n Pro_MOL_noH.ndx

6

I would also avoid using the -ur compact option when you want to avoid PBC effects. Try:
gmx trjconv -f md_0.xtc -s md_0.tpr -pbc nojump -o traj_0.xtc -n Pro_MOL_noH.ndx
gmx rms -f traj_0.xtc -s md_0.tpr -o rmsd_noH_0.xvg -tu ns -n Pro_MOL_noH.ndx

7

Thank you for you reply.
I have tried your suggested command but when i used
gmx rms command with new traj_0.xtc as input, my program error (like picture).

I have tried to make a matching run input file with gmx convert-tpr but it is still wrong

Please correct me.

image
image1090×294 11.7 KB

8

I would recommend to output the whole system when you do gmx trjconv. Otherwise you will have to make a new tpr file with only the molecules that you have written to the xtc file. You should be able to do that with convert-tpr, but I think it’s making things more complex than necessary.