yang li wrote:
Hi All,
We have several substrate-soaked data with resoltuion from
3.1~3.6A, after refinement
with the native structure, the rfree are in the range of 0.29~0.34. We
found in the expected
active site there do have extra densities, but not very clear. In some
chains the densities
are big while in other chains very small. For some good extra
densities can fit the substrate
in but cannot specify the accurate atom positions. And the map
densities seem not very convictive(low resolution or bad refinement?).
This is a relative big structure, I think it would be difficult to
rebuild it in such low resolution. We also has tried to use the sulfur
anomalous signal in the substrate in 1.54A, but it failed since all
the sulfurs in the model have no signal. So anyone has some methods
to make the density better or make sure whether it has substrate
soaked in?
Thanks!
I assume you already tried composite omit maps in CNS - if not that
would be a possibility.
If you're in desperate mode you can enhance your density magically by
using the program GraphEnt, depending on the number of reflections you
have you might have to recompile the program. In one of my trickier
cases I was able to actually build a peptide into a map obtained through
GraphEnt and then further refined it in Refmac.
Good luck !
Juergen
P.S. alternative get better diffracting crystals - e.g. analyse your
crystal contacts and do some favourable cloning.
--
Jürgen Bosch
University of Washington
Dept. of Biochemistry, K-426
1705 NE Pacific Street
Seattle, WA 98195
Box 357742
Phone: +1-206-616-4510
FAX: +1-206-685-7002
Web: http://faculty.washington.edu/jbosch
