Re: [ccp4bb] Forcing same origin on different MR solutions

- Protein crystallography

Main steps:

- Protein purification
- Crystallisation

Special:

- Programs for crystallography
- X-ray detectors

Basic tutorials:

   - Chemistry
   - Protein
   - Peptide
   - Amino Acids

Xtal community:

- CCP4BB

CCP4bb navigation

CCP4bb <-- 1999 <-- November 1999 <-- 30 November 1999

Previous message:
Subject: Forcing same origin on different MR solutions
From: Peter Grey petgxray {- at -} GMAIL {- dot -} COM
Date: 2009-10-21

Next message:
Subject: Dear friend!
From: Peter Zwart PHZwart {- at -} LBL {- dot -} GOV
Date: 2009-10-21

Subject: Re: Forcing same origin on different MR solutions
From: James Holton JMHolton {- at -} LBL {- dot -} GOV
Date: 2009-10-21

There is a program in the CCP4 Suite called "reforigin", which might be
what you want. Since all origins are equivalent, there is no way
a-priori to force an MR program to always use the "same origin", all you
can do is compare to a reference. However, a "trick" you can play on
such programs (that use RMSD as a score) is to make a "reference" PDB
file with all atom coordinates at 0,0,0, or 0,1.0,2.0 (or perhaps the
middle of the unit cell). This will origin-shift your MR solution to be
as close as possible to that point. Clearly, there are ways for this to
go wrong if your molecule is oriented "just so" but as long as you are
careful you should be okay. The best thing is to have a
"reference.pdb", and the top-scoring MR solution will often do.

Independent of "reforigin", I also wrote a jiffy to do this:
http://bl831.als.lbl.gov/~jamesh/pickup/origins.com

You run it like this: origins.com P21 reference.pdb mr_out.pdb correlate

where "correlate" means to use calculated map correlations instead of
rmsd between atoms as the "score" for alignment. This is slower, but
good for files that have no atom names in common (like heavy atom site
constellations). If you make a PDB file of grid points in your EM map,
then the "correlate" option above should work for aligning your EM map
result to a PDB file.

-James Holton
MAD Scientist

Peter Grey wrote:
> Dear crystallographers,
>
> I try to solve a MR problem in P21 with several different structures
> (and one EM map) as search models.
> I would like all solutions to have the same origin so I could compare
> them and see their relative positions.
> I think a possible solution is to bring the center of mass of all
> models (and map) to the same point.
> Is there another, easier, solution ?
> is there a way, already after molecular replacement, to bring all
> solutions to the same origin ?
>
> Thank you for your time,
>
> Peter
>

CCP4bb navigation

CCP4bb <-- 1999 <-- November 1999 <-- 30 November 1999

Previous message:
Subject: Forcing same origin on different MR solutions
From: Peter Grey petgxray {- at -} GMAIL {- dot -} COM
Date: 2009-10-21

Next message:
Subject: Dear friend!
From: Peter Zwart PHZwart {- at -} LBL {- dot -} GOV
Date: 2009-10-21