[Dock-fans] Does prepare_amber.pl (i.e, LEaP understand capping?

[Scott Brozell]

Hi,

I think this is solely an Amber LEaP issue - once the pdb is right
the leap script used in prepare_amber.pl should handle it,
but I have not tried this.

Your last message in the reflector
Subject: Re: AMBER: Does LEaP understant capping with ACE/NME?
seems to indicate that you have a next step if not success yet;
so I'll wait.

Scott

On Tue, 21 Oct 2008, Francesco Pietra wrote:

> For reasons that would be long to detail, I would like to work (DOCK6)
> with a model where the first and last residues are capped with ACE and
> NME respectively. That for more than one stretch of standard amino
> acids. The stretches are separated from one another by a TER line
> 
> The way I have built the pdb file is not understood by
> prepare_amber.pl, i.e., by leap, as it emerges from filename.log
> (where filename is the prefix of the pdb file).
> 
> As an example with one stretch, the first amino acid:
> 
> ATOM      1  N   LEU     1    xyz
> ATOM      2  CA  LEU     1
> ATOM      3  CB  LEU     1
> ATOM      4  CG  LEU     1
> ATOM      5  CD1 LEU     1
> ATOM      6  CD2 LEU     1
> ATOM      7  C   LEU     1
> ATOM      8  O   LEU     1
> ATOM      9  HA  LEU     1
> ATOM     10  HB2 LEU     1
> ATOM     11  HB3 LEU     1
> ATOM     12  HG  LEU     1
> ATOM     13 HD11 LEU     1
> ATOM     14 HD12 LEU     1
> ATOM     15 HD13 LEU     1
> ATOM     16 HD21 LEU     1
> ATOM     17 HD22 LEU     1
> ATOM     18 HD23 LEU     1
> ATOM     19  H   LEU     1
> 
> The corresponding ACE:
> 
> HETATM 4018  C   ACE    46     xyz
> HETATM 4019  O   ACE    46
> HETATM 4020  CH3 ACE    46
> HETATM 4021 HH31 ACE    46
> HETATM 4022 HH32 ACE    46
> HETATM 4023 HH33 ACE    46
> 
> The various capping residues are NOT separated from one another by a TER line.
> 
> LEaP does not understand that there is capping (while Chimera
> understands that and reproduces perfectly the whole structure). LEaP
> adds a new atom, named H, to each beginning terminal, and "missing"
> OXT to each terminal last. Then it tries to join first ACE with last
> NME. Perhaps I could avoid the last issue by placing a TER line
> between the capping residues but before embarking in a perhaps
> hopeless avenue, I ask for experience, or insight how to configure the
> pdb file.
> 
> Without capping, the model is treated correctly by prepare_amber.pl,
> though it has charges that the true protein has not, which is likely
> to alter any docking.
>