[gmx-users] Can I align individual domains (not the whole structure) in the RMSD?

[ZHANG Cheng]

Sorry, can someone help me with this question? I think this question is very important, though it seems to be very basic. I can see in many publicationis, the authors do not explicitly state how the alighment is done for the RMSD.


Thank you!


Cheng




------------------ Original ------------------
From:  "ZHANG Cheng"<272699575 at qq.com>;
Date:  Wed, Feb 27, 2019 01:09 AM
To:  "gromacs.org_gmx-users"<gromacs.org_gmx-users at maillist.sys.kth.se>;"ZHANG Cheng"<272699575 at qq.com>;

Subject:  Can I align individual domains (not the whole structure) in the RMSD?



My protein Fab has VH, VL, CL, CH domains. Usually in the RMSD calculation, I align the whole Fab first, then use a index file to calculate the domain's RMSD:
$ echo 1 19 | gmx rms -s md_0_1.tpr -f md_0_1_noPBC.xtc -o rmsd.xvg -tu ns -n index.ndx
(Group 1 is the whole protein, Group 19 is one of the domains)


Now I was told to align the domain only, and then calculate the RMSD of that domain:
$ echo 19 19 | gmx rms -s md_0_1.tpr -f md_0_1_noPBC.xtc -o rmsd.xvg -tu ns -n index.ndx
$ echo 20 20 | gmx rms -s md_0_1.tpr -f md_0_1_noPBC.xtc -o rmsd.xvg -tu ns -n index.ndx
$ echo 21 21 | gmx rms -s md_0_1.tpr -f md_0_1_noPBC.xtc -o rmsd.xvg -tu ns -n index.ndx
$ echo 22 22 | gmx rms -s md_0_1.tpr -f md_0_1_noPBC.xtc -o rmsd.xvg -tu ns -n index.ndx
because we want to see if the domain itself has changed (e.g. unfolded) throughout the trajectories, and compare which domain deforms mostly. If the whole structure is aligned, then the RMSD of each domain will be influenced by the alignment of the whole.


I can understand this saying. But is this really fair to align each domain? Is this a common practice?
Also, in most cases, the domain with more residues would have higher RMSD if each domain is aligned.


Thank you!


Cheng