[gmx-users] Re:gmxdemo: peptide to real protein and boom

[David Mathog]

Tamas Horvath wrote:
> 
> Check out what groups you have. The values at tau_t and ref_t refer to
those
> groups.
> In the default case, in the demo, those refer to "protein" and "SOL"
groups.

I went another way on this one.  Since the values in each column
are the same (at least in the demos) I modified readir.c so that it
just filled in the missing tau_t and ref_t values with those from
the last column.  After emitting a warning of course.  Then it ran
to completion.

Which brings up the next mystery.  When ngmx runs on the output
the filter screen has 

Group-1: system,protein,protein-h,c-alpha,backbone,
mainchain,mainchain_Cb, and MainChain+H
Group-2: sidechain,sidechain-H,prot-masses,nonprotein,CA,SOL,Other

In the gmxdemo with cpeptide there was only one Group-1, which
contained all of the entries from Groups 1 and 2 except CA.

What significance does one group vs. two have?  Any?  Is it just that
the program thinks the menu is too long and splits it at 15 entries?

Now to see if the Amber FF can be convinced to work with this protein...

Thanks,

David Mathog
mathog at caltech.edu
Manager, Sequence Analysis Facility, Biology Division, Caltech