dear colleagues,
i just noticed a to me somewhat unexpected behaviour of pdbset.
as it turns out, it seems to make a difference whether i do run pdbset
all in one go, changing the cell and applying transformations at once:
pdbset XYZIN ... << eof-pdbset
cell <my slightly different new cell here>
spacegroup ...
rotate euler <my amore result>
shift fractional <my amore result>
eof-pdbset
or split it up into two steps, where i first change the cell
pdbset XYZIN ... XYZOUT tmp1.pdb << eof-pdbset1
cell <my slightly different new cell here>
spacegroup ...
eof-pdbset1
and then apply the transformations
pdbset XYZIN tmp1.pdb XYZOUT ... << eof-pdbset2
spacegroup ...
rotate euler <my amore result>
shift fractional <my amore result>
eof-pdbset2
the second option works fine, with an r-factor in refmac rb refinement
that corresponds to my amore result, while in my special case the first
option results in a pdb file that does not refine in refmac. my
impression is that the pdbset "cell" card somehow does not properly
override the original, old, cell from the pdb file throughout ... or is
this maybe an implicit feature, something to do with the order in which
pdbset processes the input pdb and input cards ???
no big problem, easily worked around ... once you are aware of it, that
is ... possibly worthy of a short warning in the documentation ?
cheers
ingo
