Hi everyone:
As some of you may know, there is an inconsistency in Amber with regards to
water molecules and pdb files. In $AMBERHOME/dat/leap/lib/solvents.cmd, a
water is built like this:
r = createResidue WAT
add r h1
add r h2
add r o
That is, the order is H1, H2, O. If you use the "savePdb" command in leap
with a water residue, this is the order you get.
However, for reasons I don't completely grok, saveAmberParm creates the order
O, H1, H2; and if you use ambpdb with such a prmtop, you also get the order
O, H1, H2.
As far as I can tell, prmtop files have always used the O, H1, H2 order, and
savepdb in LEap has always been inconsistent in this fashion (at least for
the past several versions of Amber). Also, water (because of the peculiar way
it is built) seems to be the only residue where this inconsistency shows up;
I would appreciate hearing about other similar "problems".
Clearly, this confusing situation is sub-optimal. I propose to update
all of the waters in solvents.cmd to look like this:
r = createResidue WAT
add r o
add r h1
add r h2
This will give the "prmtop"-like order. I'll do this for all water models,
and use that to update solvents.lib.
So, here is my detailed proposal:
If you know that this might cause you problems (i.e. you rely on the current
behavior in some scripts, etc.) reply to this e-mail, and I won't do anything
until we clear up the problem.
If I don't hear anything in a couple of days, I'll update the change in
amber10 CVS. (I hope most of you are starting to use Amber 10.) Then, we
will see if any problems arise in actual use.
...thanks....dac
Received on Sun May 14 2006 - 06:07:07 PDT
