I just find it a little strange that leap offers to produce mol2 files (which are a standardised file format and widely used by other simulation, graphical and modification toolsets), yet does not conform to the intended file format. However if leap does not load any information regarding explicit bond order, then naturally this is understandable.
> We *should* have tleap populate the element columns in the pdb files it
> outputs. Then OB could presumably do a pdb -> mol2 conversion without any
> need to change atom names.
This works great for standard residues, but when using non-standard residues OpenBabel has no way of knowing how to connect each atom other than based on distances, which can often produce incorrect connectivity. I've not yet found a structure that allows you to specify non-standard residue connectivity (similar to leaps .off file system) for reading PDBs.
I did not realise antechamber was capable of assigning correct bond orders - is there an example available which I can follow which applies to my situation of using non-standard residues /w PDB?
Cheers,
Nick
________________________________________
From: amber-bounces.ambermd.org [amber-bounces.ambermd.org] on behalf of David A Case [case.biomaps.rutgers.edu]
Sent: 20 March 2014 01:44
To: AMBER Mailing List
Subject: Re: [AMBER] tleap output of mol2 format incorrect bond order/atom types
On Wed, Mar 19, 2014, Browning, Nicholas wrote:
>
> What would be a useful fix is for leap to output mol2 files with the
> correct SYBYL style atom types rather than printing out amber atom
> types (Which are not compatible with structures designed to load SYBYL
> atom types...).
But tleap is a part of the Amber tool chain, and Sybyl atom types would mean
that the resulting mol2 files would be of no use to Amber.
One could add a "set default SybylTypes on" flag to tleap, to put get the
desired atom types, but (a) that seems to me to be likely to create lots
of confusion; and (b) there would still be the problem of the bond orders.
Given that there is a simple workaround (running through antechamber) that
provides both Sybyl atom types and bond order assignments, I'm not yet
convinced a change is necessary.
> replace them with their elemental names as a workaround to
> get OB to interpret it correctly (OB can then assign roughly correct
> SYBYL atom types e.g C -> C.3, again based on connectivity).
We *should* have tleap populate the element columns in the pdb files it
outputs. Then OB could presumably do a pdb -> mol2 conversion without any
need to change atom names.
...dac
_______________________________________________
AMBER mailing list
AMBER.ambermd.org
http://lists.ambermd.org/mailman/listinfo/amber
_______________________________________________
AMBER mailing list
AMBER.ambermd.org
http://lists.ambermd.org/mailman/listinfo/amber
Received on Thu Mar 20 2014 - 03:00:04 PDT
