Using structure from one dataset for refinement of another


In many cases we will have multiple versions of the same dataset - e.g. with different 2D classes selected, polishing applied or not, local ctf correction, etc.

In such cases I’ll often want to select the output of a previous refinement to do a quick WS refinement of the new dataset. Right now this is possible but a little awkward - if I select a structure with all datasets shown (but no experiment yet defined for the new dataset), and then select the new dataset, the original structure is unselected and the only option is an ab initio run.

If I create a “dummy” ab initio run for the new dataset, such that it is listed in the experiments list when not filtered by dataset, then I can select an old structure and the new dataset from this page and perform a refinement - but this seems a little convoluted.

Also, on a related note, any plans to refine dataset organization - e.g. allowing users to define relationships between datasets, based on sample or project etc?

Keep up the good work, loving cryoSPARC so far!


Hi Oli,

I guess another somewhat convoluted procedure would be uploading an .mrc file from a different dataset to the current dataset. Then you can select it as a reference model for the refinement.