[ccp4bb]: Re: [sharp-discuss] Re: Heavy Atom Negative Density

[Paul Hubbard <phubbard@post.its.mcw.edu>]

***  For details on how to be removed from this list visit the  ***
***          CCP4 home page http://www.ccp4.ac.uk         ***

Hello,

In followup to my previous e-mail, I have run both Fe MAD data alone, and in
combination with the Os and Hg derivatives in SHARP. The MAD data alone gives
nice positive denisty at the FeS cluster, but I still get negative density when
I add the 2 derivatives (including the FeS atoms).

NOTE: I know the Os and Hg derivatives are useable since SAS phaing using native
in-house data with PHASES, and then doing X-Fouriers shows really nice peaks in
the Fo-Fc maps, and this is confirmed using  Fe-MAD phases, and in the
isomorphous difference Patterson. I also triple checked using SOLVE - so I'm
pretty sure its right.

However, when I run the data through RESOLVE I get, what looks like on first
impressions (though I could be very wrong), a traceable map - but no FeS cluster.
Does anyone have any idea what might be wrong? Since the MAD alone phased map
looks fine I think there is nothing wrong with scaling or having bijvoet pairs
flipped.

By the way, I view the RESOLVE map in XTALVIEW by calculating a PHS file from the
RESOLVE output with the following script, and viewing the Fo*f.o.m. map (is this
wrong?):

mtz2various HKLIN resolve.mtz HKLOUT resolve.phs << eof
OUTPUT USER  -
'(i3,2i4,4f8.2)'
labin  FP=FP FOM=FOMM PHIB=PHIM
end

I have a seminar on Friday, and wanted to have the structure properly phased by
then. Thanks!

AGS