On Sat, May 14, 2016 at 9:49 AM, <hannes.loeffler.stfc.ac.uk> wrote:
> Hi,
>
> looks like a few things have changed since I last looked into parmed...
>
> How is the version number supposed to be understood now? The one in
> AmberTools 14 I have is 15.0b, the one from AmberTools 16 is 2.4.0.
>
As the pace of ParmEd development picked up and it started getting used
outside the space of just AmberTools, it became apparent that I needed some
kind of versioning that would enable people to pin specific versions.
Versions are now monotonically increasing, and you can very easily pin a
minimum version by doing something like the following:
import parmed as pmd
assert pmd.version >= (2, 4, 0), "Upgrade ParmEd"
For AmberTools 14, 15, and earlier, the version was not as reliably updated
from revision to revision, and parmed.version did not exist (which is why
the above code snippet will always work to pin a minimum version). As Hai
mentioned, the API has been changed slightly (major changes were some
consistency improvements and package renamed from chemistry to parmed).
>
> The unit of the dihedral phase appears to have changed. In the fragment
> below it seems it is now in degrees. Is that right?
>
Yes. Unfortunately, I had forgotten that my original choice of unit for
the phase angle was radians, and every other parser for torsion phase
angles used degrees instead. Since every other angle measurement in ParmEd
used degrees, the obvious choice to make the API consistent (and eliminate
bugs in cross-program conversions) was to change to using degrees for
dihedral phase angles.
ParmEd is now out of beta and its API is stable, so these growing pains
between versions should not happen anymore.
All the best,
Jason
--
Jason M. Swails
_______________________________________________
AMBER mailing list
AMBER.ambermd.org
http://lists.ambermd.org/mailman/listinfo/amber
Received on Sat May 14 2016 - 20:00:03 PDT
