AMBER Archive (2005)

Subject: Re: AMBER: Antechamber/formatting question

From: David A. Case (case_at_scripps.edu)
Date: Sat Feb 05 2005 - 13:17:24 CST


On Fri, Feb 04, 2005, Kara Di Giorgio wrote:

> I'm trying to model a minor groove binding agent bound to a 10-mer DNA
> strand. I created the DNA in nucgen and am planning on creating a
> custom unit consisting of the guanine/compound bound together and then
> modifying the DNA-pdb to insert the custom unit instead of the guanine.
> As the DNA strand is symmetrical, one compound will be on each strand.
> (This seemed to be the method suggested in the tutorial "Simulating a
> Solvated Protein that Contains Non-Standard Residues".)
>
> 1. Is this a reasonable approach to take? Is there another way to do
> this that is easier?

yes.

>
> 2. I'm trying to create the custom unit using antechamber. I have a
> previous (very old) pdb file that has the guanine/compound already
> covalently attached. I've tried to delete all other information from
> the pdb and then run it through antechamber. It fails when trying to
> assign bond types. I get the message:
>
> Running: /usr/local/amber/exe/bondtype -i ANTECHAMBER_BOND_TYPE.AC0 -o
> ANTECHAMBER_BOND_TYPE.AC -f ac -j full
>
> Cannot successfully assign bond type for this molecule, please :
> (1) double check the structure (the connectivity) and/or
> (2) adjust atom valence penalty parameters in APS.DAT, and/or
> (3) increase MAXVASTATE in define.h and recompile bondtype.C

Leave out the "-j full" flag to bondtype; equivalently, add the "-j 5" flag to
the antechamber command.

You have two residues in your input pdb file. This is OK, but note that the
output files created will merge everything into a single residue: antechamber
really only works on single residues.
>
>
> and then it goes on and tries to run divcon where it fails due to no
> convergence.

You have a "dangling" phosphate group at the end of your molecule. Divcon
(and antechamber) need a "full" molecule, with all valencies filled. You
may also need to tell antehcamber the overall charge of your molecule, if it
is not neutral.

Note that you are trying a very big system (90 atoms). Even with correct
input, it may be tricky to get SCF convergence. If so, you may have to split
the molecule into two parts, and run each separately.

I would suggest leaving off the sugar-phosphate part of the molecule
(replacing C1' with a hydrogen, or a methyl group), and use the standard
Amber parameters for that. And you sure have an unusual chemistry here!
Be sure to check that antechamber is giving reasonable results, especially
around N14.

...good luck...dac

> p.s. I'm running on a MacG4 PowerBook through an X11 terminal window

It doesn't look like these are Mac-specific problems: I got the same behavior
you did on my Windows machine.

-- 

================================================================== David A. Case | e-mail: case_at_scripps.edu Dept. of Molecular Biology, TPC15 | fax: +1-858-784-8896 The Scripps Research Institute | phone: +1-858-784-9768 10550 N. Torrey Pines Rd. | home page: La Jolla CA 92037 USA | http://www.scripps.edu/case ================================================================== ----------------------------------------------------------------------- The AMBER Mail Reflector To post, send mail to amber_at_scripps.edu To unsubscribe, send "unsubscribe amber" to majordomo_at_scripps.edu