AMBER Archive (2007)

Subject: Re: AMBER: Prepin error (?) in xleap

From: Junmei Wang (junmwang_at_yahoo.com)
Date: Fri Aug 31 2007 - 15:22:07 CDT


I took a look at the attached pdb file and found the structure is not good enough. Since no bond connectivity information is read in for a pdb format, antechamber tries to predict the bond connectivity table itself based on the atomic distances. If the input structure is not good enough, errors may happen. For your molecule, the distance between O30 and O32 is too small and antechamber wrongly assumes there is a bond there. This is my suggestion:

(1) Try to use mol2 or sdf files as input
(2) If you really want to use pdb format, try "-j 0" flag to check unexpected bond connectivity in ANTECHAMBER_AC.AC if error happens.
Command: antechamber -fi pdb -fo mol2 -i gau3.pdb -o gau3.mol2 -j 0

For you molecule, you will find a bond is formed between O30 and O32.

Then manually delete that bond and read ANTECHAMBER_AC.AC as input (ac format)

Command: antechamber -fi ac -fo mol2 -i ANTECHAMBER_AC.AC -o gau3.mol2

Good luck

Junmei

run antechamber with "-j 0" and check ANTECHAMBER_AC.AC file to find unexpected bond connection
antechamber -fi pdb -fo mol2 -i input.pdb -o input.mol2 -j 0

Lili Peng <lilipeng_at_gmail.com> wrote: Hi Dr. Case,

Okay, I added the hydrogens to get the PDB file:

       REMARK Accelrys Discovery Studio PDB file
   
       REMARK Created: Thu Aug 30 13:40:47 Pacific Daylight Time 2007 ATOM 1 N GLU 1 3.326 1.548 -0.000 1.00 0.00 N ATOM 2 CA GLU 1 3.970 2.846 -0.000 1.00 0.00 C ATOM 3 CB GLU 1 3.577 3.654 1.232 1.00 0.00 C ATOM 4 CG GLU 1 4.267 4.996 1.195 1.00 0.00 C ATOM 5 CD GLU 1 3.874 5.805 2.429 1.00 0.00 C ATOM 6 OE1 GLU 1 4.595 5.679 3.454 1.00 0.00 O ATOM 7 OE2 GLU 1 2.856 6.542 2.334 1.00 0.00 O ATOM 8 C GLU 1 5.486 2.705 -0.000 1.00 0.00 C ATOM 9 O GLU 1 6.009 1.593 -0.000 1.00 0.00 O ATOM 10 C18 GLU 1 1.933 1.409 -0.374
 1.00 0.00 C ATOM 11 O20 GLU 1 2.167 0.156 -0.632 1.00 0.00 O ATOM 12 C22 GLU 1 0.785 1.600 -1.382 1.00 0.00 C ATOM 13 C23 GLU 1 -0.602 0.995 -1.094 1.00 0.00 C ATOM 14 C24 GLU 1 -1.750 1.187 -2.103 1.00 0.00 C ATOM 15 C25 GLU 1 -3.136 0.582 -1.815 1.00 0.00 C ATOM 16 O30 GLU 1 -4.020 1.377 -2.341 1.00 0.00 O ATOM 17 O32 GLU 1 -3.796 -0.414 -1.923 1.00 0.00 O ATOM 18 N34 GLU 1 -1.877 2.648 -2.420 1.00 0.00 N ATOM 19 HT GLU 1 3.887 0.681 0.280 1.00 0.00 H ATOM 20 HA GLU 1 3.642 3.359 -0.904 1.00 0.00 H ATOM 21
  HB1 GLU 1 2.497 3.801 1.241 1.00 0.00 H ATOM 22 HB2 GLU 1 3.878 3.116 2.131 1.00 0.00 H ATOM 23 HG1 GLU 1 5.347 4.849 1.186 1.00 0.00 H ATOM 24 HG2 GLU 1 3.966 5.535 0.297 1.00 0.00 H ATOM 25 HC GLU 1 6.112 3.597 -0.000 1.00 0.00 H ATOM 26 1H22 GLU 1 0.642 2.675 -1.495 1.00 0.00 H ATOM 27 2H22 GLU 1 1.123 1.177 -2.328 1.00 0.00 H ATOM 28 1H23 GLU 1 -0.459 -0.079 -0.981 1.00 0.00 H ATOM 29 2H23 GLU 1 -0.940 1.419 -0.149 1.00 0.00 H ATOM 30 H24 GLU 1 -1.429 0.576 -2.946 1.00 0.00 H ATOM 31 1H34 GLU 1 -0.960 3.003 -2.843 1.00
 0.00 H ATOM 32 2H34 GLU 1 -2.089 3.191 -1.522 1.00 0.00 H TER 33 GLU 1
   
   
   
       END
   
   
   
   
   
   
Still, when I try to run the command " antechamber -fi pdb -i glu3.pdb -fo mol2 -o glu3.mol2" in Amber, I get the same old error message

"Warning: the assigned bond types may be wrong...

Error: cannot run "usr/local/apps/amber9/exe/bondtype -i ANTECHAMBER_BOND_TYPE.AC0 -o ANTECHAMBER_BOND_TYPE.AC -f ac -j full" in judgebondtype() of antechamber.c properly, exit"

I double checked the bond types of each atom, and I don't see any errors. What am I doing wrong now? I have attached my updated PDB file for your convenience.

Thanks,
Lili

On 8/27/07, David A. Case <case_at_scripps.edu> wrote: On Mon, Aug 27, 2007, Lili Peng wrote:
>
> Thanks for your reply. I tried running antechamber initially (as you
> prescribed), but my converted mol2 file is really weird. The carbon atoms
> get converted to Californium and Cerium atoms. There's even a Neon atom in
> the structure, as well as many "Unknowns". I have no idea how the carbon
> atoms got converted into them.

Here is the "glu3.pdb" file:
REMARK Accelrys Discovery Studio PDB file
REMARK Created: Mon Aug 27 13:56:43 Pacific Daylight Time 2007
ATOM 1 N GLU 1 3.326 1.548 -0.000 1.00 0.00 N
ATOM 2 CA GLU 1 3.970 2.846 -0.000 1.00 0.00 C
ATOM 3 CB GLU 1 3.577 3.654 1.232 1.00 0.00 C
ATOM 4 CG GLU 1 4.267 4.996 1.195 1.00 0.00 C
ATOM 5 CD GLU 1 3.874 5.805 2.429 1.00 0.00 C
ATOM 6 OE1 GLU 1 4.595 5.679 3.454 1.00 0.00 O
ATOM 7 OE2 GLU 1 2.856 6.542 2.334 1.00 0.00 O
ATOM 8 C GLU 1 5.486 2.705 -0.000 1.00 0.00 C
ATOM 9 O GLU 1 6.009 1.593 -0.000 1.00 0.00 O
TER 10 GLU 1
HETATM 11 C A 1 -1.198 -0.215 0.736 1.00 0.00 C
HETATM 12 C A 1 0.108 0.559 0.560 1.00 0.00 C
HETATM 13 C A 1 1.275 -0.423 0.476 1.00 0.00 C
HETATM 14 C A 1 2.565 0.342 0.301 1.00 0.00 C
HETATM 15 O A 1 3.623 -0.259 0.212 1.00 0.00 O
HETATM 16 N A 1 - 1.398 -1.128 -0.431 1.00 0.00 N
HETATM 17 C A 1 -2.351 0.756 0.820 1.00 0.00 C
HETATM 18 O A 1 -3.054 0.784 1.816 1.00 0.00 O
HETATM 19 O A 1 - 2.590 1.543 -0.148 1.00 0.00 O
TER 20 A 1
END

Antechamber is limited in the types of pdb files it can handle:

1. Every atom in a residue needs to have a unique atom name. You have five
atoms all named "C". (Actually, antechamber takes care of this, but other
parts of amber will not, so it is a good idea to make sure of it by hand.)

2. The input pdb file must have all atoms present, *including hydrogens.*
This is where you really go wrong, since there are no hydrogen atoms present
in your input file. See if discovery studio can put these on, or build them
by hand in xleap, or in some other modeling program.

 3. The things you think are Californium, neon, etc, are really atom types,
not elements. Given that it had no hydrogens, antechamber tried to make
multiple bonds between lots of things, and the quantum optimization was of
course very weird.

So: edit the names in the pdb file, and get some hydrogens on there!

...good luck...dac

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

 

       
---------------------------------
Need a vacation? Get great deals to amazing places on Yahoo! Travel.
-----------------------------------------------------------------------
The AMBER Mail Reflector
To post, send mail to amber_at_scripps.edu
To unsubscribe, send "unsubscribe amber" to majordomo_at_scripps.edu