Hi Dr Scheres,
I have two questions, one concerning postprocessing, and the other about masking in refinement.
When I run post-processing after refinement in Relion, all the output mrcs are plain black, and the final resolution is unrealistically high. And I think the map looks pretty clean in Chimera with threshold at 0.045. So I wonder what else I can try.
`which relion_postprocess` --i Refine3D/run1/run1_ct24 --o Refine3D/run1/postprocess --angpix 1.3 --auto_mask --inimask_threshold 0.045 --extend_inimask 3 --width_mask_edge 3 --auto_bfac --autob_lowres 10
== Perform auto-masking ...
+ density threshold: 0.045
+ extend ini mask: 3 pixels
+ width soft edge: 3 pixels
== Extending initial binary mask ...
1.15/1.15 min ............................................................~~(,_,">
== Making a soft edge on the extended mask ...
1.05/1.05 min ............................................................~~(,_,">
== Masking input maps ...
+ randomize phases beyond: 17.2138 Angstroms
== Applying sqrt(2*FSC/(FSC+1)) weighting (as in Rosenthal & Henderson, 2003) ...
== Fitting straight line through Guinier plot to find B-factor ...
+ fit from resolution: 10
+ fit until resolution: 0
+ slope of fit: -nan
+ intercept of fit: -nan
+ correlation of fit: -nan
== Low-pass filtering final map ...
+ filter frequency: 2.6
== Writing out put files ...
+ Processed map: Refine3D/run1/postprocess.mrc
+ Processed masked map: Refine3D/run1/postprocess_masked.mrc
+ Auto-mask: Refine3D/run1/postprocess_automask.mrc
+ Metadata file: Refine3D/run1/postprocess.star
+ EMDB xml-format FSC curve: Refine3D/run1/postprocess_fsc.xml
+ FINAL RESOLUTION: 2.6
The other question is about masking. I am now working on a double-disc like structure with a D8 symmetry, but I suspect that there is some inter disc flexibility that is dragging down the resolution, so I think I would try to mask one disc and do a C7 refinement. So my question is, how to create such mask, can I do it in Relion or I need other program?
Thanks a lot.
Best Regards,
Tat
|