B-factor for exposure filter

Submitted by kmanikandan on Wed, 2015-09-02 08:19

Forums

Unblur & Summovie

Hi,

I have a FalconII movie data. total dose: 40 e/A2 in 30 frames.

I have used the unblur to align and exposure filter the frames with the default parameters (Bfactor of 1500). I have calculated the power specturm of the averaged (summed) images before and after the unblur filtering.I see that the high resolution Thon-rings are more visible for the original image than for the exposure-filtered image. Is it expected like this?. Please help me to understand this.

Do I have to give the correct Bfactor for the filter? In that case, as i do not have the 3d reconstruction, how do i get the Bfactor?

with kind regards

Mani.

timgrant Wed, 2015-09-02 08:31

Hi Mani,

Yes in general you should see that the Thon-rings are less visible after the exposure filter, as the filter is removing signal in the sum (it is removing high frequency signal from the later frames, for the protein this is a good thing as it is removing radiation damage induced noise). For this reason, it can be better to use a non filtered sum for the CTF estimation, and the filtered sum for everything else, you can do this by running unblur without the exposure filter to create the sum for CTF estimation, and then using summovie to make the filtered sum using these alignment parameters. If the effect on your Thon rings is minimal however, and you still see a good number of rings in your filtered sums, it is likely fine to just use them of the CTF estimation.

If you have restore noise power set to YES, then there is no need to sharpen the sum (or no more need than the original sum at any rate). The Bfactor parameter is a low pass filter used during the alignment, it is not applied to the final sum.

Cheers,

Tim