AMBER Archive (2006)

Subject: Re: AMBER: molsurf.c error

From: Miguel Ferreira (tetrass_at_hotmail.com)
Date: Fri May 05 2006 - 06:46:24 CDT

Thanks Dr. Case!
I added the line above in the molsurf.c file following the line 1055:
   if( sarg1 < 0.0 )
      printf( %8.3f %8.3f %8.3f %8.3f\n", ai->rad, aj->rad, probe_diam,
d_ij);
I don´t know exactly what would be expected to obtain with this change but I
did not see any difference in my output files. I just added the line... Did
I miss something?

The com.all.out and rec.all.out files contain the corrected reaction field
energy and surface area but in the lig.all.out file the surface area is
missing. So, for some reason molsurf is not able to calculate the surface
area for the ligand.

complex.com.all.out
BOND = 596.9050 ANGLE = 1855.8336 DIHED = 8880.8291
VDWAALS = -9260.9542 EEL = -74191.8043 HBOND = 0.0000
1-4 VDW = 3181.9073 1-4 EEL = 38929.9730 RESTRAINT = 0.0000
corrected reaction field energy: -11291.709316
surface area = 33653.618

protein.rec.all.out:
BOND = 552.7191 ANGLE = 1663.5701 DIHED = 8621.9749
VDWAALS = -8729.0500 EEL = -73803.2483 HBOND = 0.0000
1-4 VDW = 3051.8491 1-4 EEL = 40780.9593 RESTRAINT = 0.0000
corrected reaction field energy: -11341.989783
surface area = 35859.669

ligand.lig.all.out
BOND = 44.1859 ANGLE = 192.2635 DIHED = 258.8542
VDWAALS = -79.7698 EEL = 868.2059 HBOND = 0.0000
1-4 VDW = 130.0583 1-4 EEL = -1850.9864 RESTRAINT = 0.0000
corrected reaction field energy: -2160.686578

After create pqr files for complex, ligand and protein, I ran molsurf in
each of them. The results were:

-- > molsurf complex.pqr 1.4

15504 atoms read

WARNING: 4 cusps intersection: not trimmed (surface area will be slightly
overestimated
WARNING: 3 cusps intersection: not trimmed (surface area will be slightly
overestimated)
WARNING: 3 cusps intersection: not trimmed (surface area will be slightly
overestimated)
WARNING: 5 cusps intersection: not trimmed (surface area will be slightly
overestimated)
WARNING: 3 cusps intersection: not trimmed (surface area will be slightly
overestimated)
WARNING: 4 cusps intersection: not trimmed (surface area will be slightly
overestimated
WARNING: 3 cusps intersection: not trimmed (surface area will be slightly
overestimated)
WARNING: 4 cusps intersection: not trimmed (surface area will be slightly
overestimated
surface area = 32572.773

-- > molsurf protein.pqr 1.4

15040 atoms read

WARNING: 4 cusps intersection: not trimmed (surface area will be slightly
overestimated
WARNING: 3 cusps intersection: not trimmed (surface area will be slightly
overestimated)
WARNING: 3 cusps intersection: not trimmed (surface area will be slightly
overestimated)
WARNING: 5 cusps intersection: not trimmed (surface area will be slightly
overestimated)
WARNING: 3 cusps intersection: not trimmed (surface area will be slightly
overestimated)
WARNING: 4 cusps intersection: not trimmed (surface area will be slightly
overestimated
WARNING: 3 cusps intersection: not trimmed (surface area will be slightly
overestimated)
WARNING: 4 cusps intersection: not trimmed (surface area will be slightly
overestimated
surface area = 34948.509

-- > molsurf ligand.pqr 1.4

464 atoms read

surface area = 3320.178

I am confused with these results. What the warnings mean? Isn´t it strange
that running molsurf alone, the ligand surface area can be computed but not
when running the mm_pbsa script? On the other hand, the surface area values
obtained for complex and protein are a little bit different from those
listed in the their all.out files. Is this normal?

The calculations were done with amber8. I am sending the complex.pqr as
attachment.
Any suggestion would be very very helpful.
Thanks a lot
Miguel

>From: "David A. Case" <case_at_scripps.edu>
>Reply-To: amber_at_scripps.edu
>To: amber_at_scripps.edu
>Subject: Re: AMBER: molsurf.c error
>Date: Thu, 4 May 2006 13:21:28 -0700
>
>On Thu, May 04, 2006, Miguel Ferreira wrote:
>
> > However I am still getting the error: “molsurf: molsurf.c:1055:
>is_buried:
> > Assertion `sarg1 >= 0.0' failed”. Has anyone experienced this problem
> > before?
>
>This is what the Google box at the top of the Amber web page is designed
>to help with. Searching on "sarg1" brings up the following suggestion:
>
> http://amber.ch.ic.ac.uk/archive/200410/0209.html
>
>Try to strip your problem down to as small a pqr file as you can, and see
>if
>you get any good ideas. If you can post the problem file, others can look
>at
>it....without that, we can't reproduce the problem, and won't be easily
>able
>to fix it, either.
>
>...dac
>
>-----------------------------------------------------------------------
>The AMBER Mail Reflector
>To post, send mail to amber_at_scripps.edu
>To unsubscribe, send "unsubscribe amber" to majordomo_at_scripps.edu

_________________________________________________________________
MSN Busca: fácil, rápido, direto ao ponto. http://search.msn.com.br


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