AMBER Archive (2006)

Subject: Re: AMBER: VMD PDB into LEAP Readable PDB

From: David A. Case (
Date: Thu Oct 05 2006 - 10:13:41 CDT

On Wed, Oct 04, 2006, Akshay Patny wrote:
> I have generated a GPCR protein system embedded in membrane
> using the VMD scripts. The system now has:
> 1. Protein 2. POPC bilayer 3. Water 4. Chloride Ions
> I intend to do the MD simulation in Amber8/9. Now the
> problem is that this PDB file is not in a format which is
> readable by LEAP properly.

Sounds like you will have to do some editing. You might ask on the VMD list
about this. I will note that LEaP follows PDB standards pretty
closely: if you have atom and residue names that are too long, this
violates the pdb standard itself, and not just LEaP.

> 1. The names of some heavy atoms are 4 characters long e.g.
> C321. I guess this is not allowed as the max characters
> allowed are 3 except for hydrogens which can have 4.

The pdb standard states that the atomic symbol must be right-justfied in
columns 13 and 14. Hence, a carbon must have a "C" in column 14. You can
have a name like "1C32" (columns 13-16), and Amber will convert that to
"C321". But I'm not sure whether an arbitrary digit is allowed in column 13
or not.... Generating three character names is certainly the safer way to

> 4. Is chain information required?

No, but you will need to have a TER card between every pair of chains.


The AMBER Mail Reflector
To post, send mail to
To unsubscribe, send "unsubscribe amber" to