Dear Mousheng,
as others, I'm also a bit confused about the steps you did:
On Sun, Oct 17, 2010 at 02:29:12PM -0500, Wu, Mousheng wrote:
> Dear All,
>
> I have a MAD dataset at 4Å and shelxD can find clear-cut 10
> solutions (my protein has 12 methionines).
Good.
> I ran autosharp to refine them followed by density modification.
Yes.
> After density modification, I ran solvent flattening.
Confused ... what program/procedure did you use for what you call
"density modification" and what for "solvent flattening"?
> Then I calculated the anomalous map using the phase from sharp and
> the electron density map from solvent flattening.
More confusion: what do you mean with "anomalous map"? One might call
a map with anomalous differences and some phases such an "anomalous
map" ... but I guess you rather mean an electron density map with
phases from some procedure based on anomalous differences, right?
So you calculated an electron density map from SHARP: which map
coefficients did you use? It should be:
fft hklin eden.mtz mapout eden.map <<e
LABI F1=FB PHI=PHIB
e
How did you calculate the electron density map after density
modification? If you did density modification within the
SHARP/autoSHARP system it should be
fft hklin eden_flat_53.2pc.mtz mapout eden_flat_53.2pc.map <<e
LABI F1=FBshasol PHI=PHIBshasol
e
Note: do _not_ use any figure-of-merit column when claculating maps!
Nearly all programs will give you map coefficients (amplitude and
phase) that are already correctly weighted!
Note-2: SHARP/autoSHARP outputs slightly different columns - depending
if you want a map of the light-atoms-only or onw that shows the
average HA contribution as well. As a rule: use FB*/PHIB* for MAD/SAD
and Fcent*/PHIcent* for SIRAS/MIRAS.
> The anomalous map shows the density around these 10 selenium sites
> are clear and round.
Good.
> However, the density map from solvent flattening showed that only 4
> selenium sites have clear and round density.
Is the map after "density modifcation" or the one after "density
modification plus solvent flattening"? It is very unclear ...
> The density around these 4 sites clearly showed beautiful
> helices. surprisingly, other 6 selenium sites have poor density or
> no density at all. The electron density around them is not very
> good and the predicted helical density is flat. I check the electron
> density before I ran solvent flattening. The result is same. I am
> quite confused about the big difference from these two maps.
We first need to establish what steps you performed and how to see if
these are the actually correct maps to look at.
> I also try SnB to find the selenium sites. The solutions are same as
> those from ShelxD. But How to explain the poor density around the
> selenium sites in the density modification map? Is there any
> problem with my selenium sites?
Remember that the purely HA-phased maps (i.e. directly after SHARP)
will be dominated by those heavy atoms ... so no surprise they are
nice and round Se density.
Se-MET (or S-Met for that matter) often has alternate conformations in
the side-chain: maybe your phases are actually better and therefore
model those side-chains now more accurately, i.e. as disordered
side-chains. If your environment around those sites is also disorderd
('flattened' helices) this is all not surprising.
Cheers
Clemens
--
***************************************************************
* Clemens Vonrhein, Ph.D. vonrhein AT GlobalPhasing DOT com
*
* Global Phasing Ltd.
* Sheraton House, Castle Park
* Cambridge CB3 0AX, UK
*--------------------------------------------------------------
* BUSTER Development Group (http://www.globalphasing.com)
***************************************************************
