AMBER Archive (2004)

Subject: Re: AMBER: RESP vs. AM1-BCC charges for isoprene

From: Joe Harriman (s808o_at_unb.ca)
Date: Sat Jul 17 2004 - 07:44:47 CDT


Hi Oliver,

Two things that might be of interest to you. First, like Rhonda said make sure that you
tell antechamber the right charge during the AM1-BCC calculation. In addition, I have
also experimented with AM1-BCC charges. I have compared the charges obtained from
AM1-BCC to several of my molecules with RESP charges which had been previously been
calculated. Overall I noticed that there was too much charge deviation (ie. +/- 0.5 to
1.0 charge unit deviation on individual atoms), especially in larger molecules.
However, I am dealing with peptides of 100+ atoms. Therefore, it would have been
useful for me to use AM1-BCC as it could save literally weeks of calculation time. The
AM1-BCC method may work better with smaller molecules with fewer functional groups.
Overall, the take home message that I got from this is that although AM1-BCC is much
faster than RESP it is not as accurate. Hope this helps.

Joe Harriman

-------------------------------
David J. Harriman
Computational Chemistry
Dept. of Chemistry
Univeristy of New Brunswick
Toole Hall, Room 228
j.harriman_at_unb.ca
-------------------------------

Quoting Oliver Hucke <ohucke_at_u.washington.edu>:

> Dear Amber users,
>
> I have derived partial charges for an isoprene moiety with AM1-BCC and
> with the resp procedure. I was surprised by the big differences of some
> charge values - as you can see in this sketch:
>
> AM1-BCC RESP
>
> H3 0.04 H3 0.10
> | |
> C -0.06 H 0.11 C -0.36 H 0.09
> \ / \ /
> -0.12 C=====C -0.17 0.2 C=====C -0.26
> / \ / \
> C C -0.06 C C -0.24
> | | | |
> H3 H3 0.04 H3 H3 0.09
>
> Especially the carbons engaged in the double bond look suspicious to me
> in the resp output (change of one charge from -0.12 with am1-bcc to 0.2
> with resp). Might this be a resp fit artefact?
>
> My understanding was is the AM1-BCC method was designed to reproduce the
> RESP charges. So, I am wondering if something went wrong here.
>
> Your comments are highly appreciated.
>
> Oliver
>
> =====================================
> Details of what I did are given here:
>
> AM1-BCC:
>
> antechamber -fi pdb -i isopren.pdb -fo mol2 -o isopren.mol2 -c bcc
>
> RESP:
>
> 1. Geometry optimization with gaussian98:
> # HF/6-31G* opt scf=tight test
> 2. ESP calculation:
> # HF/6-31G* pop=mk iop(6/33=2) iop(6/42=6) geom=check
>
> Then:
>
> antechamber -fi gout -i isopren.out -fo ac -o isopren.ac
> espgen -i isopren.out -o isopren.esp
> respgen -i isopren.ac -o isopren.respin1 -f resp1
> respgen -i isopren.ac -o isopren.respin2 -f resp2
> resp -O -i isopren.respin1 -o isopren.respout1 -e isopren.esp \
> -t iso.qout1
> resp -O -i isopren.respin2 -o isopren.respout2 -e isopren.esp \
> -q iso.qout1 -t iso.qout2
>
>
>
> --
> _______________________________________________________________
>
> Oliver Hucke, Dr.
> Health Sciences Building - K418C
> University of Washington 1959 NE Pacific St.
> Dept. of Biochemistry phone: (206) 685 7046
> Box 357742 fax : (206) 685 7002
> Seattle, WA 98195-7742 email: ohucke_at_u.washington.edu
> _______________________________________________________________
>
> -----------------------------------------------------------------------
> The AMBER Mail Reflector
> To post, send mail to amber_at_scripps.edu
> To unsubscribe, send "unsubscribe amber" to majordomo_at_scripps.edu
>

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