[ccp4bb]: Re:

["Eleanor J. Dodson" <ccp4@ysbl.york.ac.uk>]

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

sameeta bilgrami wrote:
> 
> ***  For details on how to be removed from this list visit the  ***
> ***          CCP4 home page http://www.ccp4.ac.uk         ***
> 
> dear all,
>      I am trying to solve a structure through MIR. I
> have got 3 derivatives Hg Th and Ur. All the
> derivatives as well as the native have data upto 2.5 A
> resolution.
>      I tried SOLVE. The FOM for the Phases finallly
> obtained is 0.48 till 3.74 A resolution. The overall
> Phasing power for the Hg derivative is 0.91, for Th it
> is 0.21 and that for Ur it is 0.28.The final map that
> i get after running solomon has an sfall reliability
> index of 0.31 till 2.5 A resolution.


 It is very important all your derivatives are on the same origin and
hand.

You dont say what your space group is, but many have perfectly valid
alternate origins. There is a CCP4 tutorial 2000 explainingthis - you
should follow it..

You use the phases from the Hg to do difference fouriers to position the
U and Th. Then you can be sure you have the same origin..

Have you done this? If not it might explain the very low FOMs for these
two.

Use your centric data if possible to do preliminary refinementr of the
sites from all 3 derivatives, and cross check for consistency.

 And do you have anomalous data?

>      I am new to the method of MIR (it is being done
> for the first time in our lab). I have the following
> doubts that i know only a novice can have. I hope you
> will all bear with me:
> 1) I have read people saying that they can see the
> protein and solvent boundary clearly or they can see
> the outline of two domains. But in O the maps can be
> seen upto a certain radii mostly upto 8 A radii. How
> do we see the entire Electron density together in
> O...will it help if i just increase the radii of the
> map that has to be drawn...or is there another program
> to visualize the whole electron density of an
> assymetric unit together.


 Use mapslicer of the CCP4 GUI automatic plotting of your maps to look
at blocks of sections..

> 2) From what i have read people first start putting in
> sec. str. elements (mostly helices)...and model the
> loops only in the end. But my protein has no helices
> and very little B-sheet and mostly loops. The nmr
> model of a similar protein shows that even the
> B-sheets dont have standard ramachandran values. What
> will be the best way to start modeling this protein. I
> am seeing certain portions in the density which seem
> like continuos strand of a protein. Should i start
> putting something there.
> 3) The phasing power for Hg is just Ok and that for Ur
> and Th is bad. Is phasing power the property of the
> derivative as such or it depends on the program that
> has found out the sites and refined them. Even the FOM
> are  not good. On what does FOM and R-cullis depend.


 FOM and Rcullis described in log files or in tutorial 2000.

> 4) The ccp4 has a documentation for all the programs
> with detailed descriptions of the keywords used in the
> program. Is there a similar manual for understanding
> the logfiles.

And look at the loggraphs - also available through GUI..

See above..

Best wishes Eleanor