AMBER Archive (2009)

Subject: Re: [AMBER] ptraj_hbond_problem

From: Thomas Cheatham (
Date: Fri Jun 05 2009 - 00:28:11 CDT

> trajin tmd1.mdcrd 1 2
> trajout origihbondtry.out
> donor mask :1-40_at_O
> acceptor mask :1-40_at_N : 1-40_at_H
> hbond series h1 out hbond.dat includeself distance 3.6 angle 160 out

OK, in previous posts you successfully chose atoms but didn't find any
distances. Here there is a seg fault which provides very little
information and nothing for me to go on...

> \-/ Version: "AMBER 9.0 integrated" (April 2006)

Use the latest AmberTools version at http//

> PTRAJ: donor mask :1-40_at_O
> Mask [:1-40_at_O] represents 1260 atoms

You found 1260 atoms. This must be the water. I would have used the
selection :WAT_at_O

> PTRAJ: acceptor mask : 1-40_at_N :1-40_at_H
> Segmentation fault

This suggests there is a mismatch in N-H or that N-H do not exist, but I
am guessing here. Ah, looking through my DNA, I do not see atoms named
"N". N1 or N9 perhaps?

ptraj is complaining because it doesn't recognize the mask. Probably also
since : 1 is not the proper syntax (no spaces!).

Previously you found no waters; let's try something simple.

Take residue #1. Do "rdparm prmtop" where prmtop is your parameter
topology file. Type "atoms :1". Pick two that are connected for the
acceptor command, perhaps

acceptor mask :1_at_O5' :1_at_H5T

Set the distance to be large and the angle to be -1. If even this doesn't
find waters, look at the structure and make sure waters haven't drifted
away. If they have, you need to image (see the archives).

As mentioned on the list many times previously, experiment. Try to
decipher what is going wrong. N-H is not present in DNA. What atoms do
you want to see interactions to?


AMBER mailing list