[phenixbb] Trouble with occupancy refinement

Christian Roth christianroth034 at gmail.com
Fri Feb 5 10:34:02 PST 2016


Hi Chris, what would concern me is that one glucose have atoms with
different occupancies or all of them. That is chemically unlikely. You
should define sensible constraint groups, I think.

Cheers Christian
On 5 Feb 2016 18:09, "Chris Fage" <cdfage at gmail.com> wrote:

> Hi Pavel,
>
> The B-factors of these atoms are quite high (most are 100+), so maybe the
> occupancy is compensating as you said. After refinement, the maps don't
> look much different. Perhaps it's not worth bothering with occupancy
> refinement, then?
>
> I think you got my question. I was asking if it is unusual that only 5/93
> occupancies were refined to values other than 0 or 1.
>
> Best,
> Chris
>
> On Fri, Feb 5, 2016 at 6:38 PM, Pavel Afonine <pafonine at lbl.gov> wrote:
>
>> Hi Chris,
>>
>> I've built several glucose molecules from my cryosolution into regions of
>>> weaker electron density. Attempting to refine the occupancies of the
>>> corresponding atoms, I set the value of each to 0.25 in the pdb file.
>>> However, after running phenix.refine, most of the atoms still have
>>> occupancies of 1.00:
>>>
>>
>> perhaps they are fully occupied but highly mobile which explains weak
>> density? Or B-factor refines to larger value so that occupancy tries to
>> compensate? How maps look like after refinement?
>>
>> 14 atoms with occ = 0.00
>>> 5 atoms with 0.00 < occ < 1.00
>>> 74 atoms with occ = 1.00
>>>
>>> Is this unusual?
>>>
>>
>> Do you mean is it unusual to have 5 atoms in your structure with
>> occupancies between 0 and 1? Hm.. I don't know, may be not! I think I'm not
>> understanding what's the real question here..
>>
>> Pavel
>>
>>
>
> _______________________________________________
> phenixbb mailing list
> phenixbb at phenix-online.org
> http://phenix-online.org/mailman/listinfo/phenixbb
> Unsubscribe: phenixbb-leave at phenix-online.org
>
-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://phenix-online.org/pipermail/phenixbb/attachments/20160205/7a0a51f3/attachment-0001.htm>


More information about the phenixbb mailing list