AMBER Archive (2003)

Subject: Re: Antechamber questions

From: Scott Shandler (shandles_at_mail.med.upenn.edu)
Date: Mon Mar 31 2003 - 17:42:35 CST

Thanks for your response, I do have a follow-up question though. (see
below)

On Fri, 28 Mar 2003, David A. Case wrote:

>_On Thu, Mar 27, 2003, Scott Shandler wrote:
>_>
>_>
>_> 1) I am running Amber 7 with Mopac 7 and have linked the appropriate
>_> files so that I can finally get output file, however I get the following
>_> message:
>_>
>_> >~/work/seq/test ->antechamber -i hetero_antechamber.pdb -fi pdb -o
>_> >hetero.prepin -fo prepin -c bcc
>_> >
>_> >Running: /usr/local/amber7/exe/mopac.sh ANTECHAMBER_MUL.MOP
>_> >ANTECHAMBER_MUL.OUT
>_> >Mopac job ANTECHAMBER_MUL.MOP started
>_> >Mopac job ANTECHAMBER_MUL.MOP has finished
>_> >
>_> >Warning, 45 bond types may not be correctly assigned
>_
>_sounds bad...does your molecule have a good geometry (before, but especially
>_after) the MOPAC run? Does the output from mopac look good...check for
>_any indications of failures.
>_

I checked both geometry before and after the MOPAC run and all seems to be
intact. The calculations do appear to complete, and I think I might have
a lead on which "bond types may not be correctly assigned". The "peptide"
that I am attempting to determine parameters for consists of both natural
and non-natural backbone atoms (i.e. benzene ring embedded in the
backbone) Upon inspection of the ANTECHAMBER_BCC_TYPE.AC file, the bond
records that antechamber is complaining about seem to be the ones that are
non-natural in order of occurance. Would it be safe to assume that the
warning message is simply indicating that the non-natural backbone (and
side chain) bonds that MOPAC determines are unable to be confirmed
(looked up and assigned) by Antechamber since it is not naturally occuring
peptide?

BOND 7 4 5 1 N H
BOND 8 4 6 1 N C !!!
BOND 9 6 7 1 C C !!!
BOND 10 6 13 1 C C !!!
BOND 11 7 8 1 C C !!!
BOND 12 8 9 1 C C !!!
BOND 13 8 14 1 C C !!!
BOND 14 9 10 1 C O

>_>
>_> 2) I am attempting to generate molecular fragment parameters with Amber 7,
>_> and was wondering if there is any way to generate backward compatible
>_> united atom parameters?
>_
>_No, we have not done anything on the united atom front for many years.

Thanks, we were due for an upgrade anyway ;)

>_
>_..good luck...dac
>_
>_

Thank you for your time,

-SJS