AMBER Archive (2005)

Subject: Re: AMBER: RED-II failure to recognize GAMESS output

From: FyD (fyd_at_u-picardie.fr)
Date: Sat Sep 03 2005 - 02:33:38 CDT

Quoting Cenk Andac <cenk_andac_at_yahoo.com>:

> I have been using Red-II to determine RESP charges
> for the NMeacetamide (NMA) example in RED-II manual
> (p26). I used GAMESS to optimize coordinates for NMA.
> I also used the following set-up in RED-II for the
> optimization process.
>
> $XRED = "Off";
> $NP = "1";
> $QMSOFT = "GAMESS";
> $OPT_Calc = "On";
> $MEPCHR_Calc = "On";
> $CHR_TYP = "RESP";
> $CHR_VAL = "0";
> $MLT_VAL = "1";
> $TITLE = "nacet";
> $DIR = "~/dir";
> $MOL_START = "~/dir/nacet.pdb";
> $JOB_OPT = "~/dir/JOB1-gam1.log";
>
> In principal, GAMESS minimized NMA and printed
> out JOB1-gam1.irc and JOB1-gam1.dat files in gamess'
> src directory.

For using GAMESS & R.E.D. please see the R.E.D.-II manual, pages 7-8.
i. e. 'Remarks regarding the use of "GAMESS" with R.E.D.'
Just in case ;-)

> Here are some words in ~/dir/JOB1-gam1.log file:
> JOB1-gam1.log:
> ************************************
> 1 ***** EQUILIBRIUM GEOMETRY LOCATED *****
> COORDINATES OF ALL ATOMS ARE (ANGS)
> ATOM CHARGE X Y Z
> ----------------------------------------------------
> C 6.0 2.0074569211 1.0062470751 0.0000000000
> H 1.0 1.6249586848 -0.0033937275 0.0000000000
> H 1.0 1.6413208735 1.5288656361 0.8781704037
> H 1.0 1.6413208734 1.5288656362 -0.8781704038
> C 6.0 3.5222773671 0.9566272924 0.0000000000
> O 8.0 4.1288339649 -0.0788038913 0.0000000000
> N 7.0 4.1252034162 2.1701985420 0.0000000000
> H 1.0 3.5523591717 2.9823246679 0.0000000000
> C 6.0 5.5597011361 2.3455464160 0.0000000000
> H 1.0 6.0184115984 1.3701706433 0.0000000000
> H 1.0 5.8850779964 2.8876758549 -0.8816339325
> H 1.0 5.8850779964 2.8876758549 0.8816339325
> [...]
>
> ----------------------------------------
> CPU timing information for all processes
> ========================================
> 0: 454.256942 + 19.266071 = 473.523013
> ----------------------------------------
> ddikick.x: exited gracefully.
> ----- accounting info -----
> Fri Sep 2 20:23:49 PDT 2005
> ...
> ***************************************
>
> Although NMA was optimized by GAMESS,
> RED-II gave me the following error message.
>
> * Software checking *
> rungms [ OK ]
> gamess.01.x [ OK ]
> ddikick.x [ OK ]
> resp [ OK ]
> The structure/conformation(s) is/are being optimized
> [ FAILED ]
> See the file "JOB1-gam1.log"
> Press Enter to exit

I think the way you use R.E.D. is correct...

There is a trick anyway somewhere, please see
http://www.u-picardie.fr/labo/lbpd/RED/FAQ-II.htm#2
to understand the GAMESS||Gaussian format recognized by R.E.D.
- The author names are checked to differentiate GAMESS and Gaussian output
(since, with R.E.D.-II, geom. opt. output from Gaussian can be used in MEP
computation with GAMESS, and vis-versa)
- EQUILIBRIUM GEOMETRY LOCATED is checked for GAMESS
- EXECUTION OF GAMESS TERMINATED NORMALLY is also checked for GAMESS...

I just did a test with MeOH and it worked fine. What is the GAMESS version you
use ? The version I used is "GAMESS VERSION = 22 NOV 2004 (R1)"

> I then re-ran RED-II this time by turning off
> "OPT_calc" option and using JOB1-gam1.log as input
>
> $XRED = "Off";
> $NP = "1";
> $QMSOFT = "GAMESS";
> $OPT_Calc = "Off";
> $MEPCHR_Calc = "On";
> $CHR_TYP = "RESP";
> $CHR_VAL = "0";
> $MLT_VAL = "1";
> $TITLE = "nacet";
> $DIR = "~/dir";
> $MOL_START = "~/dir/nacet.pdb";
> $JOB_OPT = "~/dir/JOB1-gam1.log";
>
> This time, RED-II gave the following error
> message :
>
> * Selected QM Software *
> GAMESS
> * Selected optimization output *
> GAMESS
> Invalid optimization OUTPUT !
> Press Enter to exit.
>
> Could anyone of you please let me know why
> RED-II does not recognize the GAMESS output
> (JOB1-gam1.log)?

Same rules (as above) are applied here with this new one:
- The atom order (& _chemical elements_) in the initial PDB file & in the QM
output must be the same.

Finally, you could try the following variables:
 $OPT_Calc = "Off";
 $MEPCHR_Calc = "On";
 $DIR = "~/dir";
 $MOL_START = "./nacet.pdb";
 $JOB_OPT = "./JOB1-gam1.log";

Indeed, at the end of the R.E.D. job "dir" is renamed into "dir-1", and your new
directory which contains your charges is "dir" (see the 'Remark' in the
R.E.D.-II manual page 27).

Tell me if it works now, regards, Francois 

-- 
 • F.-Y. Dupradeau •
DMAG EA 3901 & Faculte de Pharmacie, Amiens, France
       ••••
http://www.u-picardie.fr/labo/lbpd/FyD.htm
-----------------------------------------------------------------------
The AMBER Mail Reflector
To post, send mail to amber_at_scripps.edu
To unsubscribe, send "unsubscribe amber" to majordomo_at_scripps.edu