[phenixbb] high B Factor after real space refinement against a cryoEM map

Pavel Afonine pafonine at lbl.gov
Thu Oct 17 00:44:04 PDT 2019


Hi,

> I am trying to do the real space refinement for my model against a 5.0 
> angstrom cryoEM map. All parameters look pretty good except the B 
> factor. The mean of the B factor is close to 400. I am just wondering 
> whether the low resolution makes the B factor this high.
>
> Does anyone have any idea or suggestion on the high B factor during 
> real space refinement? Is it possible to reduce it? What is the actual 
> meaning of this B factor?
>

given how much cryo-EM maps are 'massaged' along the way, B factors are 
not necessarily very meaningful. They are refined though to provide the 
best match between model and cryo-EM maps to give you realistic 
map-model correlations.
Apart from this, refining Bs at this resolution may result in large 
numbers to compensate for the lack of density for side chains, for example.

Bottom line, I don't think this is something to worry about.

Pavel

-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://phenix-online.org/pipermail/phenixbb/attachments/20191017/84ce5eb4/attachment.htm>


More information about the phenixbb mailing list