Lowpass filter in Volume Tools doesn't output the filtered map

Troubleshooting
1

Hi all!

Someone suggested to me that I could lowpass filter my map to get an idea of what filter resolution I should use in 3D classification.

I’m trying to use Volume Tools in v4.6 to accomplish this as described here: About Low pass filtering of maps - #6 by mmclean

The only parameter I changed was the lowpass filter, which I set to 20 A as a test. Although the job itself shows a clearly lowpass filtered map in the events log, the actual .mrc file is identical to the input map. Are there more steps to the procedure than just simply changing the lowpass filter parameter, or could this be a bug?

Edit: After some testing I found that this behavior isn’t specific to local filtering but also occurs with cropping in real space as well. Although the plot in the event log looks fine, the actual map that gets output is identical to the input map.

Best,
cbeck

2

Seems to work fine on my end (on 4.6.2) - possibly something specific to your system? I wonder if related to this post: Inconsistent Map Level Values Between Download and Local File

3

@cbeck
Please can you let us know how you accessed

Did you (a) download it through the CryoSPARC GUI, (b) view in the CryoSPARC GUI’s map viewer, or (c) copy it directly from the job directory?

4

Hi wtempel!

I originally opened the map directly from the job directory. However, after seeing your message, I tried downloading the map through the cryoSPARC GUI and saw that the map is properly lowpass filtered and cropped. Thank you! Is this the expected behavior, and is there any way to access the lowpass filtered map from the project directory?

Best,
cbeck

5

Not the expected behavior - on my system, both are correctly filtered

6

There was an issue fixed in v4.6.2 that somewhat resembles the problem you described. Does the release note match your own observation; if so, have you already updated to v4.6.2?

7

I haven’t updated to v4.6.2 yet, but I’ll update this thread once I do. This looks like it’s specific to my machine, because some of my colleagues using v4.6.1 don’t have this issue.