AMBER Archive (2004)

Subject: Re: AMBER: residue insertion

From: Scott Brozell (sbrozell_at_scripps.edu)
Date: Fri Mar 19 2004 - 17:01:02 CST

Hi,

On Wed, 10 Mar 2004, David A. Case wrote:
> On Wed, Mar 10, 2004, Yunfeng Hu wrote:
>
> > Hi, leap does not seem to recognize the residue numbering such as 27,
> > 27A, 27B... I wonder how to fix this. Thanks.

E.g., the PDB file
...
ATOM 239 OE1 GLN L 27 -6.614 5.118 63.924 1.00 34.82
ATOM 240 NE2 GLN L 27 -5.872 7.014 64.792 1.00 20.88
ATOM 243 N SER L 27A -5.631 2.964 68.743 1.00 25.00
...
ATOM 259 CD2 LEU L 27B -1.505 1.975 65.145 1.00 21.12
ATOM 260 N VAL L 27C -4.507 -2.304 66.704 1.00 11.85
...
ATOM 280 HE2 HIS L 27D -6.138 -5.907 57.786 1.00 0.00
ATOM 281 N SER L 27E -6.200 -7.045 63.802 1.00 32.36
...
ATOM 288 HG SER L 27E -5.593 -10.356 62.225 1.00 0.00
ATOM 289 N TYR L 28 -7.367 -9.869 65.801 1.00 23.96

> Someone would have to delve into the source code. A workaround is
> to use unique numbers in your input files, e.g. change " 27A" to
> " 127 ". When the residue number (in columns 23-26) changes, LEaP
> knows to start a new residue. But it still won't keep the original
> numbering scheme.

Leap was doing the right thing during the loadpdb command;
these messages indicate that new residues were being created when the
insertion code changed:
Splitting pdb file residue 27 into GLN and SER
Splitting pdb file residue 27 into SER and LEU
Splitting pdb file residue 27 into LEU and VAL
Splitting pdb file residue 27 into VAL and HIS
Splitting pdb file residue 27 into HIS and SER
 5 residues had naming problems.
 Residues split, so res numbers will not correspond to pdb.

Further evidence is the output of the describe command:
R<SER 26>
R<GLN 27>
R<SER 28>
R<LEU 29>
R<VAL 30>
R<HIE 31>
R<SER 32>
R<TYR 33>

However, if two consecutive residues with the same sequence number and
the SAME names had different insertion codes then leap would fail.
That problem has been fixed, and we are designating it a feature instead
of a bug fix. The patch file is available on request.

Note that as Dave mentioned Leap does not retain the original
residue identification (sequence number, residue name, and insertion code)
from the pdb file, ie, as above SER 27A in the pdb becomes 28 in Leap, etc.
That problem won't get fixed until Amber 9.

Scott Brozell

-----------------------------------------------------------------------
The AMBER Mail Reflector
To post, send mail to amber_at_scripps.edu
To unsubscribe, send "unsubscribe amber" to majordomo_at_scripps.edu