Single-stranded dna rmsd analysis

[imrbay]

GROMACS version:2020
GROMACS modification: Yes
Hi,I applied md simulation using amber force field which is OL15,on my single-stranded apo form DNA structures.for analysis I get confused .in order to rmsd analysis,I created index file each of residues (DC,DT,DA,DG) and added new groups for selection.
selection group for least squares fit I chose DA residues ,select group for Rmsd calculation System
BUT the graph is

selection group for least squares fit I chose system residues ,select group for Rmsd calculation DA,
The graph is

Could you help me which graph I can prefer or which groups I can choose ?
could anybody explain me ?

[jalemkul]

Neither. You should not fit or calculate with System, because that just gives you results that are washed out by the solvent (which is what appears to be the case in the first plot).

[imrbay]

Thank you,so two graphs are not correct

into the picture shows that my groups,so in order to rmsd calculation for DC or any groups how to do selection.Could you guide me please

[imrbay]

I created backbone groups .

when I chose firstly backbone,secondy system The graph is

When I chose firstly system,secondly backbone .The graph is :

What is my fault,what is the difference between two graphs ,which graph is true for me ,could anybody explain and suggest to me,I will be very grateful.Thank you

[sibasankar.p]

Use the same selection from the reference structure that you are fitting for RMSD. While calculating for backbone, exclude hydrogen atoms. You can add one more extension to your command -fit rot+tans. It seems you system is not equilibrated properly. Try to extend the run.