Reference model restraints in real space refinement?
Hi all, Is it possible to use reference model restraints at the moment during real space refinement (in phenix.real_space_refine)? I can’t see it in the documentation. On a related note, would it be possible at some point to allow users to set d_max in the phenix.map_to_structure_factors utility? At the moment it seems like the user can set d_min but not d_max - sometimes it is convenient to restrict the low res to 20 or 50Å, rather than the whole box size of an EM-map, which is the default. Best, Oliver.
Hi Oliver,
Is it possible to use reference model restraints at the moment during real space refinement (in phenix.real_space_refine)? I can’t see it in the documentation.
no, sorry.
On a related note, would it be possible at some point to allow users to set d_max in the phenix.map_to_structure_factors utility?
It is possible in current Phenix: phenix.map_to_structure_factors map.ccp4 d_min=2 Note, d_min cannot be arbitrary as it needs to satisfy sampling requirements. No d_max though as I could not imagine a valid reason for cutting the low-res reflections. It's easy to add but I would like to understand why this is necessary before I do it. I think whether you get a box or a sphere with reflections is governed by the "box" keyword that is False by defalut (so that by default you get a sphere). Isn't it the case? Pavel
On 04/12/2015 01:16 PM, Pavel Afonine wrote:
Hi Oliver,
> No d_max though as I could not imagine a valid reason for cutting the low-res reflections. It's easy to add but I would like to understand why this is necessary before I do it.
>
> I think whether you get a box or a sphere with reflections is governed by the "box" keyword that is False by defalut (so that by default you get a sphere). Isn't it the case?
>
I think the "box" here refers to the artificial orthorhombic unit cell in which the structure factors are calculated- so that if there is no d_max limit, the lowest resolution reflections are the first-order reflections which with resolution equal to the cell edges, i.e. "the box".
participants (3)
-
Edward A. Berry -
Oliver Clarke -
Pavel Afonine