The number of inputted and outputted particles in the extract from micrographs job doesn’t add up. I added around 1.5 million particles from two inspect picks jobs, but the output is less than their sum, and even the number of exposures is incorrect in the output as well. In the ‘Inputs’, the numbers show correctly, but there’s something wrong with the output groups.
I put it 17633 micrographs and a total of around 3.4 mil particles but only ~2.4 mil are extracted. Am I misunderstanding something or doing something wrong?
It will automatically skip any particle where you can’t extract the entire box from the data because the box goes beyond the edge of the micrograph. Since you have a large box (448pix), many are skipped. I would also think it might be useful to have an option to “pad” the data with noise and generate a second output with all particles, but this is fairly complex and utility is low. If I’m right, you can use a tiny box, your particle count will go up, and the only difference in plots (or inspect picks comparison) is new particles at the edges.
I think the output exposures from extract automatically removes any exposures from which no particles were extracted…?
@CryoEM2 is right about the particle extraction, which is a feature I both like and dislike at the same time… it’s really useful for small things, but costs me valuable particles when working with larger objects. It would be nice to have a toggle option for “CryoSPARC Classic” extraction or “RELION style barcode edge”.
