Re: [ccp4bb] How to determine water number?

[Paul Emsley]

AntonioLeung wrote:
 
I am a novice working on protein structure. When I pick water using 
COOT, too many waters picked, filling in the whole cell.


Too many - hmm. Did you, by any chance, peak-search a difference map 
without first changing the default sigma cut-off to (say) 3.5?


Paul.

[ccp4bb] How to determine water number?

[AntonioLeung]

Hi all,
  
 I am a novice working on protein structure. When I pick water using COOT, too 
many waters picked, filling in the whole cell. My question is how can I 
determine which water is needed, which is not needed?

Re: [ccp4bb] How to determine water number?

[Pavel Afonine]

Hi Antonio,

My question is how can I determine which water is needed, which is not 
needed?


may be this will give you some clues - here are the approximate criteria 
for water picking that are used in automatic water picking and 
refinement in  phenix.refine:


1) peak at mFo-DFc map is higher than ~3sigma, and
2) peak center is within a hydrogen bond to another atom (water or 
macromolecule), and
3) peak has approximately the same shape as a water molecule would have 
at this resolution and local environment, and

4) peak at 2mFo-DFc is higher than ~1.5 sigma,

after a round of coordinate and B-factor refinement,

the criteria (2-3-4) are still ok, and

5) refined B-factor of newly placed water is meaningful (didn't jump to 
large value), otherwise a water is deleted.


There are a few other technical tricks to make this process robust and 
efficient at high resolution, higher than ~1.2A or so, but this is, I 
guess, beyond of what you were asking about.


Pavel.