On Fri, Apr 10, 2009, Joe Krahn wrote:
> I think it would be nice to include a %SIZE flag before the %FORMAT
> flag, and maybe create a more complete I/O routines to replace the
> simple nxtsec.f. However, any improvements in PRMTOP handling depend on
> whether AMBER still wants compatibility with the old "raw" formatted
> PRMTOP files.
For now, we still want/need to support the old, "raw" format.
>
> Or, if AMBER developers are satisfied with a fairly primitive file
> format, maybe it is sufficient just to define a strict 80-character
> limit for all PRMTOP lines.
I'm inclined to require the 80 character limit, and to beat Mark on the
head with a wet noodle, after which he can revised the CHARMM section to
match the spec.
Note that changing the prmtop specs would require changes to many
external programs (such as VMD), in addition to whatever changes would
be needed in Amber itself. We probably don't even know all the programs
that read prmtop files. So, we should be conservative here, and require
a really good reason for any changes.
...dac
p.s. Joe: I now have a collaboration with a crystallographer who has
about a dozen chains in her structures. Do you have a procedure that
would (a) update a prmtop file to include original residue numbers and
chain ID's, (b) allow ambpdb to use such information?...thanks!
_______________________________________________
AMBER-Developers mailing list
AMBER-Developers.ambermd.org
http://lists.ambermd.org/mailman/listinfo/amber-developers
Received on Sun Apr 12 2009 - 01:08:54 PDT
