*** For details on how to be removed from this list visit the ***
*** CCP4 home page http://www.ccp4.ac.uk ***
Dear Francisco
I have found Kevin
Cowtan's CCP4 Fffear program most excellent for
searching maps for missing stuff. He has some
good examples in the documentation. You can use a
refmac 2fo-fc map to search but Fffear expects a
FOM. So I have started using FWT and PHIWT with a
dummy FOM column of all 1.0s (I add mine in
SFTOOLS). It is very important to filter the
model and the map with a 6-ish angstroem radius
as he suggests. If you want just hints where to
put the molecules then 10 degree steps is ok but
these might not refine into the density. You can
do a finer search but that will be a lot longer
searching. Make a 'solvent' mask of the current
model to exclude hits to the ones you know about.
You can use SFALL and then MAPMASK or else the
excellent MAMA. You need to apply a NOT command
in MAMA anyway to invert the logic of the map
since you want stuff that is in what you
currently think of as the 'solvent' area.
Fffear gives you a pdb with lots of possible hits
but I usually trim it to the top 26 since these
are the best and have distinct chain letters.
There was some discussion on this BB a while back
on how to go through and delete false fffear hits
by chain id. Fffear thinks all the residues are
ALA but will keep all your residue atoms in the
solutions - looks a bit odd when they belong to an RNA probe however ;)
Having said all this I am sure we will soon have
a great new version with a swashbuckling name. I
have used Fffear to search DMMULTI maps but soon
I expect I will be using 'HAWKINS' to search PIRATE maps ;)
shiver me timbers!
good luck
Martyn
Martyn F. Symmons
Oppenheimer Research Fellow
Department of Pathology
University of Cambridge
"Abair ach beagan agus abair gu math e"
Seanfhacal Gaidhlig (Gaelic Proverb).
At 15:16 06/12/2005, you wrote:
*** For details on how to be removed from this list visit the ***
*** CCP4 home page http://www.ccp4.ac.uk ***
Dear All :
I got a pretty difficult case (I guess) of Molecular Replacement. My
protein crystallize in a P21 spacegroup with cell dimensions of around
100 x 200 x 220 A. By Matthews analysis the ASU could accomodate from 16
to 18 molecules of the protein (Molecular weight around 38 kDa). The
resolution of the data is around 2.7 A.
I have run Phaser and it found 10 molecules within the ASU, but the
program failed to find more. I have made and NCS averaging with these
molecules to build up a more precise model, and tried again Phaser with
the same results. I have also tried Molrep with the same results.
Does anybody experienced a problem like this ?. Any suggestion of
strategies to find the rest of the molecules that will complete the
ASU ?.
Any help will be sincerely appreciated.
Cheers
Francisco
--
***********************************
Francisco J. Enguita, Ph.D.
Host-pathogen interactions group
Protein Crystallography Laboratory
Av. da Republica
Estação Agronomica Nacional
2780-157 Oeiras
PORTUGAL
Phone : +351-21-4469669
Fax : +351-21-4433644
E-mail : [EMAIL PROTECTED]
Web : http://xtal.itqb.unl.pt
************************************
