resolution range to fit B-factor


What's the best resolution range to use for obtaining a good B-factor estimate (bfactor v1.04)?
In my case, the structure is 13A (0.5 FSC cutoff) or 11A (0.143 FSC cutoff).

Is there a good way to determine if the B-factor value that you are using is too high (or low)?

Thanks for the help.

Ideally, you have data beyond 10 Å resolution to minimize the influence of the shape transform of your particle/structure. The shape transform depends on the size of your particle. A small particle will produce a transform with significant power extending to higher resolution than a large particle. You should perhaps read the paper by Fernández et al. (2008). They describe the algorithm that our B-factor software uses. In this paper, the resolution range used is 10 Å - 15 Å. Please keep in mind that the measured B-factor may be biased by the shape transform of your particle. So unless you can work with data at higher than 10 Å resolution you should be careful with interpretation of the measurements.