You dont say how you solved this, but usually now the solution methods
should give the correct hand, providing you indexed the data correctly
and did not muddle up hkl and -h,-k,-l measurements!
Fromm the heavy atom search you cant tell the hand so both x,y,z in P32
and -x,-y,-z in P31 will give equally good scores.
Ditto the phasing statistics from the MAD phasing will be equally good
on both hands.
But two sets of MAD phases are not simply related by a change of sign or
adding of 180 degrees so the maps on one hand are of different quality
to maps on the other hand.
There can be exemptions when the heavy atom sites are related by a
centre of symmetry..
That means that the density modification procedures should work better
with one solution than the other..
You need to
1) check if the Zn sites are related by some operator.
2) look at the density modification statistics and see how clearly they
distinguish one hand from the other.
3) check processing - it is now very hard to get this wrong with the
standard software,but obv it can still happen..
DNA is difficult I know - the symmetry can get extremely complex..
Eleanor
Ruchi Anand wrote:
Dear All,
We are working with a DNA binding protein with 4 Zn sites in in ASU. Using SnB
and then CNS we were able to get a map using MAD phasing and could visualize
the density for the double helix of the DNA but it was a left handed spiral
instead of the usual right handed one. The space group which yielded the result
was P3(2). We tried to flip the sites and change the space group to P3(1) but
we are not able to generate a sensible map. Any advice regarding the best way
to proceed will be very helpful. The resolution is around 3.2.
Thanks
Ruchi
