Hi Karthik,
in phenix.refine there is a parameter for this:
ignore_zero_occupancy_atoms=True
It tells phenix.refine: "Yes, please, do not include atoms with zero
occupancy factors into mask calculation (that is: ignore such atoms
while computing the mask)". This is what I had in mind when introducing
this parameter. If this sounds too confusing (I don't feel it since I'm
not native English speaker), then I would appreciate if native speakers
suggest a better name.
Cheers,
Pavel.
On 9/17/09 11:32 AM, Karthik S wrote:
> How is the solvent mask handled for zero occupancy atoms?
> In general is it justified that i fear some of the electron density
> belonging to the protein in the periphery of the molecule can be
> solvent flattened and i never see them thereafter in my electron
> density maps or is this just not going to happen (i.e., the density
> for protein will never be averaged out)?
> I have a 1.97A dataset for a ~570 odd residue protein with two domains
> that had independently been solved earlier. My phaser solution had
> most of the residues fit in density for the full length protein but i
> am in the process of building the residues that were absent in the
> search models and there are some trouble areas that i set to zero
> occupancy and did a simulated annealing refinement in phenix . (Is
> this the omit map that will point to general features in the region
> that i had set to zero occupancy?)
> This from phenix documentation:
> mask
> ......
> ignore_zero_occupancy_atoms= True Include atoms with zero occupancy into
>
> mask calculation
>
> .....
>
> so the comment means the zero occupancy will be included but the
> command itself says ignore the zero occupancy atoms to be true, isn't
> the command definition and comment contradictory to each other.
>
> Please educate me...
>
> I am really sorry for the half sent email earlier.
>
> Thanks,
>
> Karthik
>
> _____________________________
> Graduate Student, Biophysics
>
> University of Michigan
>
> Ann Arbor, MI 48109
>
> 734-763-3384
>
> [log in to unmask]
>
> _____________________________
>
> On Thu, Sep 17, 2009 at 2:28 PM, Karthik S <[log in to unmask]> wrote:
>
>> How is the solvent mask handled for zero occupancy atoms?
>> In general is it justified that i fear some of the electron density belonging to the protein in the periphery of the molecule can be solvent flattened and i never see them thereafter in my electron density maps or is this just not going to happen (i.e., the density for protein will never be averaged out)?
>> I have a 1.97A dataset for a ~570 odd residue protein with two domains that had independently been solved earlier. My phaser solution had most of the residues fit in density for the full length protein but i am in the process of building the residues that were absent in the search models and there are some trouble areas that i set to zero occupancy and did a simulated annealing refinement in phenix . (Is this the omit map that will point to general features in the region that i had set to zero occupancy?)
>> This from phenix documentation:
>> mask
>> ......
>>
>>
|