[gmx-users] g_wham 4.6.7 version output query

[Justin Lemkul]

On 12/8/14 3:35 PM, Agnivo Gosai wrote:
> Hi Users
>
> I googled on the issue and found the following :
>
> https://www.mail-archive.com/gmx-users@gromacs.org/msg53884.html
>
> In the above mentioned link an user had previously wrote about the
> creation of this pmfintegrated.xvg file and as mentioned it is not
> being created all the time..
>
> I also found the following from the source code of 4.5.5. version
>
> if (opt->verbose)
>      {
>          fp=xvgropen("pmfintegrated.xvg","PMF from force
> integration","z","PMF [kJ/mol]",opt->oenv);
>          for(j=0;j<opt->bins;++j)
>              fprintf(fp,"%g  %g\n",(j+0.5)*dz+opt->min,pot[j]);
>          ffclose(fp);
>          printf("verbose mode: wrote %s with PMF from interated
> forces\n","pmfintegrated.xvg");
>      }
>
> However I am using version 4.6.7. And I also use -v in the command
> line for g_wham as indicated in the trailing mail. Can someone say
> what is going on. Lookwise the  pmfintegrated.xvg is so smooth that I
> am tempted to use it , but I don not know why and how it is generated.
>

Without looking deeply into the code, it's probably an "ideal" PMF, or in theory 
what your output should look like.  You have clear sampling defects and you 
should not use what is effectively debug output in place of your real results. 
Run a few more windows in areas lacking sampling and you'll get something you 
can actually use.

-Justin

-- 
==================================================

Justin A. Lemkul, Ph.D.
Ruth L. Kirschstein NRSA Postdoctoral Fellow

Department of Pharmaceutical Sciences
School of Pharmacy
Health Sciences Facility II, Room 629
University of Maryland, Baltimore
20 Penn St.
Baltimore, MD 21201

jalemkul at outerbanks.umaryland.edu | (410) 706-7441
http://mackerell.umaryland.edu/~jalemkul

==================================================