Reverse downsampling?


Is there any easy way to select the original particles corresponding to a downsampled set, without re-extracting?

I have to run 3D-VA with downsampled particles due to memory constraints, but after clustering I want to get the original particles and refine to high resolution. Is this possible? It seems like something one ought to be able to do with the low level results interface, but I can’t see exactly how.


1 Like

Hi @olibclarke you should be able to do this simply by connecting the output particle group from your clustering output to the next job (refinement), and then expand the connected group in the job builder, and then find the original extraction at full box (doesn’t have to be a subset of particle, can be the whole dataset) and then from the output tab of that job, connect just the particle “blob” result group and override the “blob” in the job builder.
The refinement job will use only the intersection of all result groups that are in the particles set, so it will still keep the cluster of particles but use the original raw particle data.

Hope that helps!

1 Like

Ah I see, that works! Thanks @apunjani!