Nothing in the half-maps after refinement 2.0

3D Refinement
1

Dear colleagues,

I faced a similar issue with the half-maps.
Nothing in the half-maps after refinement - Troubleshooting - CryoSPARC Discuss

They are nothing but dust.

image
image686×641 18.3 KB

Any suggestion is appreciated.

Thank you.

Kind regards,
Dmitry

2

Updates - that happens when using 3D Homogeneous refinements and Local resolution refinement.

The half-maps in NU-refinement is ok.

Cryosparc 4.4.1 + 240110

Kind regards,
Dmitry

3

@Dmitry Please can you post additional information:

  1. The filename of the map that you showed in your first post.
  2. The output of the following command, with the correct project and job IDs specified, for the job that produced that map:
    cryosparcm cli "get_job('P912', 'J934',  'job_type', 'version', 'params_spec', 'input_slot_groups')"
1 Like
4

Hello @wtempel,

  1. cryosparc_P13_J128_013_volume_map.mrc
  2. Please see the output of the following command

cryosparcm cli "get_job('P13', 'J128', 'job_type', 'version', 'params_spec', 'input_slot_groups')"

cryosparc_user@cryoem1:~$

cryosparcm cli "get_job('P13', 'J128', 'job_type', 'version', 'params_spec', 'input_slot_groups')"

{'_id': '663bacc09d2d40135037ff92', 'input_slot_groups': [{'connections': [{'group_name': 'particles_0', 'job_uid': 'J109', 'slots': [{'group_name': 'particles_0', 'job_uid': 'J109', 'result_name': 'blob', 'result_type': 'particle.blob', 'slot_name': 'blob', 'version': 'F'}, {'group_name': 'particles_0', 'job_uid': 'J109', 'result_name': 'ctf', 'result_type': 'particle.ctf', 'slot_name': 'ctf', 'version': 'F'}, {'group_name': 'particles_0', 'job_uid': 'J109', 'result_name': 'alignments3D', 'result_type': 'particle.alignments3D', 'slot_name': 'alignments3D', 'version': 'F'}, {'group_name': 'particles_0', 'job_uid': 'J109', 'result_name': 'motion', 'result_type': 'particle.motion', 'slot_name': None, 'version': 'F'}, {'group_name': 'particles_0', 'job_uid': 'J109', 'result_name': 'location', 'result_type': 'particle.location', 'slot_name': None, 'version': 'F'}]}], 'count_max': inf, 'count_min': 1, 'description': 'Particle stacks to use. Multiple stacks will be concatenated.', 'name': 'particles', 'repeat_allowed': False, 'slots': [{'description': '', 'name': 'blob', 'optional': False, 'title': 'Particle data blobs', 'type': 'particle.blob'}, {'description': '', 'name': 'ctf', 'optional': False, 'title': 'Particle ctf parameters', 'type': 'particle.ctf'}, {'description': '', 'name': 'alignments3D', 'optional': True, 'title': 'Particle 3D alignments (optional)', 'type': 'particle.alignments3D'}], 'title': 'Particle stacks', 'type': 'particle'}, {'connections': [{'group_name': 'volume', 'job_uid': 'J111', 'slots': [{'group_name': 'volume', 'job_uid': 'J111', 'result_name': 'map', 'result_type': 'volume.blob', 'slot_name': 'map', 'version': 'F'}, {'group_name': 'volume', 'job_uid': 'J111', 'result_name': 'map_sharp', 'result_type': 'volume.blob', 'slot_name': None, 'version': 'F'}, {'group_name': 'volume', 'job_uid': 'J111', 'result_name': 'map_half_A', 'result_type': 'volume.blob', 'slot_name': None, 'version': 'F'}, {'group_name': 'volume', 'job_uid': 'J111', 'result_name': 'map_half_B', 'result_type': 'volume.blob', 'slot_name': None, 'version': 'F'}, {'group_name': 'volume', 'job_uid': 'J111', 'result_name': 'mask_refine', 'result_type': 'volume.blob', 'slot_name': None, 'version': 'F'}, {'group_name': 'volume', 'job_uid': 'J111', 'result_name': 'mask_fsc', 'result_type': 'volume.blob', 'slot_name': None, 'version': 'F'}, {'group_name': 'volume', 'job_uid': 'J111', 'result_name': 'mask_fsc_auto', 'result_type': 'volume.blob', 'slot_name': None, 'version': 'F'}, {'group_name': 'volume', 'job_uid': 'J111', 'result_name': 'precision', 'result_type': 'volume.blob', 'slot_name': None, 'version': 'F'}]}], 'count_max': 1, 'count_min': 1, 'description': '', 'name': 'volume', 'repeat_allowed': False, 'slots': [{'description': '', 'name': 'map', 'optional': False, 'title': 'Initial volume raw data', 'type': 'volume.blob'}], 'title': 'Initial volume', 'type': 'volume'}, {'connections': [{'group_name': 'mask', 'job_uid': 'J111', 'slots': [{'group_name': 'mask', 'job_uid': 'J111', 'result_name': 'mask_refine', 'result_type': 'volume.blob', 'slot_name': 'mask', 'version': 'F'}]}], 'count_max': 1, 'count_min': 0, 'description': '', 'name': 'mask', 'repeat_allowed': False, 'slots': [{'description': '', 'name': 'mask', 'optional': False, 'title': 'Static mask', 'type': 'volume.blob'}], 'title': 'Static mask', 'type': 'mask'}], 'job_type': 'homo_refine_new', 'params_spec': {'refine_ctf_global_refine': {'value': True}, 'refine_num_final_iterations': {'value': 2}, 'refine_scale_min': {'value': True}}, 'project_uid': 'P13', 'uid': 'J128', 'version': 'v4.5.0'}

5

Hi @Dmitry! For the jobs with the empty half-maps, does the full map have anything in it, or is it also empty?

1 Like
6

hello @rposert yes, the full map is ok. :slight_smile:

7

Thanks for the info @Dmitry, that’s interesting…

Could you please share a bit more info about the half maps?

  • It would be interesting to see the voxel histogram for one of the half maps and for the full map (like below)
    image

  • If you contour down your half map, does the noise fill up the entire box (like below)? If so, can you see anything inside the box using the Graphics > Side View panel to “cut into” it?

    image
    image912×765 13.4 KB

8

Hello, I want to proceed with the discussion. Now, it appears that the local refinement after particle extraction does not have the half-maps; it is just the same empty dusty area. The NU refinements with the same input seem to work ok.
Any ideas?

Thank you

9

Hi @Dmitry, welcome back!

I have a few questions about your Local Refinement:

  • How large is the region included in your mask?
  • What does the GSFSC curve look like?
  • Does your full map (i.e., the map download) look like?
  • If you filter the half maps (by running, say, volume gaussian #1 sdev 4 where #1 is the id of the half map in ChimeraX) do you see anything appear?
1 Like
10

Thank you, I am glad to be with you all again :)!

  1. size - the complex contains 2 almost equal halves (the box size is 600x600), so one is removed.
  2. GSFSC
    image
    image420×215 16 KB
  3. Do you mean the result of the local resolution - map? it looks ok
  4. If I do that - volume gaussian #1 sdev 4, it appeared!!!
    So please advise - what can we conclude?
  • Are the halfmap-resutls ok, just not visible?
  • can those results be safely deposited on the wwPDB OneDep System website?

Thank you!

11

Ah, yes, this is actually a general feature of half maps! The further your GSFSC resolution is from the Nyquist frequency (twice the pixel size, at the far right of the GSFSC plot), the noisier they will look. This effect is not a cause for concern regarding deposition.

What’s happening here is that your map only has reliable signal out to 3.3 Å, but it has noise at all frequencies out to Nyquist (which is <1.3 Å in your case). In the half maps, all of this noise appears at full strength, so your map gets swamped out in the detail. When you Gaussian filter the map, that high-frequency noise goes away, leaving only your map!

The local refinement map you download (not the half maps) is made by

  1. Averaging together the two half maps
  2. Filtering the maps by the GSFSC (in effect similar to the Gaussian filter you applied, but specifically tuned to your data instead of an arbitrary value)

This process gets rid of the high-frequency noise, helping your map appear just like the Gaussian filter made your half map appear.

One final note: when your reconstructions are this far away from Nyquist, you may want to consider downsampling them using Downsample Particles. This reduces the box size of the particles which significantly speeds up jobs, and as long as Nyquist remains above your GSFSC resolution there is no significant drawback to doing so.

3 Likes
12

Many thanks!

Kind regards,
Dmitry

1 Like