Re: [AMBER] calculate ligand RMSD for MD simulations; nofit or not?

From: David A Case <david.case.rutgers.edu>
Date: Mon, 22 Feb 2016 12:21:32 -0500

On Mon, Feb 22, 2016, Marc Zhang wrote:
>
> One of the amber tutorials online suggests that "nofit" should be used when
> calculating the RMSD for the ligand but not for the protein.

I think (hope) that thinking carefully about what "nofit" means will help:

"nofit" implies that you compute the coordinate difference between A and B
using just the coordinates you are given.

Without the "nofit" option, the rms command will first superimpose the A and B
structures, and then compute the coordinate difference.

> rms reference mass :1-289

This command superimposes the current protein structure with that of the
reference structure. The ligand goes along for the ride.
>
> rms reference mass out ligand_rmsd.dat :290 nofit

Now this command gives the difference between the ligand coordinates in the
reference and those in the (rotated and translated) postion coming from the
first rms command.

If you left the "nofit" out of the second command, you would be superimposing
the ligand on the reference ligand, with no regard to the protein postions.
(The fact that you had done the first (protein) rms command would be
irrelevant.)

Try doing it both ways, and visualize the results. That's often the best
way to work through something like this.

...dac


_______________________________________________
AMBER mailing list
AMBER.ambermd.org
http://lists.ambermd.org/mailman/listinfo/amber
Received on Mon Feb 22 2016 - 09:30:04 PST
Custom Search