AMBER Archive (2009)

Subject: Re: [AMBER] MM - PBSA problem

From: Cenk \(Jenk\) Andac (cenk_andac_at_yahoo.com)
Date: Mon Jan 26 2009 - 10:20:25 CST

Hi Musa,

Your files work fine at TR-GRID.
However, I looked at file $AMBERHOME/src/mmpbsa/mm_pbsa_statistics.pm
at TR-GRID and I noticed that I commented out lines 132, 133 and 137 as follows.

line
132 # "ELRAELE" => [1,0,'ELRAELE = +(-?\d+\.\d+)'],
133 # "EPB" => [1,0,'EPB = +(-?\d+\.\d+)'],
.
.
137 # "PBNONPOL" => [1,0,'ENPOLAR = +(\d+\.\d+)'],

I think this is how I got MMPBSA working at TR-GRID in the past. I am not really sure though if lines 132,133 and 137 are essential parts of the PB section. Nevertheless, you don't need these lines for the @PB parameters in binding_energy.mmpbsa file you sent. Please somebody corrects me if I am mistaken here.

I also cross-checked your outputs using $AMBERHOME/src/mmpbsa/mm_pbsa_statistics.pm from AMBERv9 and I exactly got the same result as compared to that of MMPBSAv10 (with the modifications suggested above).

cheers,

Jenk

musa özboyacı <musaozboyaci_at_gmail.com> wrote: Dear Jenk,
I attach the necessary files that you ask me and the log file. Hope you can
find out what the problem is!
By the way, I don't use TR-GRID!

All the best,

Musa

2009/1/26 Cenk (Jenk) Andac

> Hi Musa,
>
> Could you do the following things for me?
>
> 1) please specifiaclly list here all @PB parameters you use in your
> mmpbsa.in file to run your mmpbsa jobs.
>
> 2) rerun mmpbsa only for 2 snaphosts of the complex or receptor or ligand
> species..and send me $PREFIX_com.all.out or $PREFIX_rec.all.out or
> $PREFIX_lig..all.out (where $PREFIX is the prefix you use)
>
> I'll run a few test jobs at the TR-GRID facility of Turkiye to see if we
> get similar error messages. If we don't, then i would recommend that you
> replace all mmpbsa source codes in your $AMBERHOME/src/mmpbsa directory with
> fresh ones and rerun mmpbsa.
>
> 3) I hope that you are not currently using TR-GRID for your MMPBSA exps
> because I have been currently (and temporarily) modifying some mmpbsa codes
> for some users at TR-GRID to be able to run it in parallel. If you are
> using TR-GRID for MMPBSA jobs , please let me know...
>
>
> cheers,
>
> Jenk
>
>
>
>
>
>
> musa özboyacı wrote: Dear Jenk,There appears
> nothing wrong with the output, and it did not give
> any error. Indeed. The "ELRAELE" problem was occuring while statistics
> calculations. I guess there is something wrong in statistics code and the
> same problem does not happen in GB calculations.
>
> Best regards,
>
> Musa
>
> 2009/1/25 Cenk (Jenk) Andac
>
> > Hi Dave,
> >
> > "David A. Case" wrote:
> >
> > Just a extra note here: cutcap, xcap, etc. *are* in the &cntrl namelist
> > for sander, but not for pbsa. I don't know which program mm_pbsa is
> > calling with this input.
> >
> > Bottom line: the original error may have a very different origin than
> > the one I suggested.
> >
> > ...dac
> >
> > Thanks for reminding me about using sander in MMPBSA v10.
> > I have MMPBSA versions 8 through 10 installed on my computer and I
> > sometimes forget what version goes with what.
> >
> > So, Musa may want to retry the previous pbsa.in using SANDER v10 as
> > follows:
> >
> > sander -O -i pbsa.in -o pbsa.out -p prmtop -c snapshot1.x
> >
> > I wonder now how pbsa.out will come out.
> >
> > cheers,
> >
> > jenk
> >
> >
> > _______________________________________________
> > AMBER mailing list
> > AMBER_at_ambermd.org
> > http://lists.ambermd.org/mailman/listinfo/amber
> >
> _______________________________________________
> AMBER mailing list
> AMBER_at_ambermd.org
> http://lists.ambermd.org/mailman/listinfo/amber
>
>
>
> _______________________________________________
> AMBER mailing list
> AMBER_at_ambermd.org
> http://lists.ambermd.org/mailman/listinfo/amber
>
_______________________________________________
AMBER mailing list
AMBER_at_ambermd.org
http://lists.ambermd.org/mailman/listinfo/amber

       
_______________________________________________
AMBER mailing list
AMBER_at_ambermd.org
http://lists.ambermd.org/mailman/listinfo/amber