Dear community,
Does anyone know whether the “loose mask” in the non-uniform refinement job correspond to the “mask_refine” in the job output?
Thanks!
Kate
Hi @sk.huang,
The “mask_refine” output corresponds to the dynamic mask (or static, if inputted) used to suppress the solvent in the corners/edges of the box. This mask can be affected by changing the dynamic masking parameters, or by inputting a mask into the job. It is a different mask from the masks in the FSC plot.
In the refinement FSC plots, the “loose mask” curve uses an auto-generated mask with a falloff of 15 Å. For all but the last FSC plot, the “tight mask” curve uses an auto-generated mask with a falloff of 6 Å; this tight mask is output as ..._mask_fsc.mrc so it can be manually inspected. In the last FSC plot of the job, the tight mask used for FSC computation is further auto-tightened, and is output as ...mask_fsc_auto.mrc; this is the same mask used for the corrected FSC as well.
The “Validation (FSC)” job allows for more careful FSC computation and we generally recommend using this job to generate FSC curves, since you can manually input your own mask (which may be the ...mask_fsc, ...mask_fsc_auto, or a manual mask from volume tools.
Best,
Michael
5 Likes