Why do duplicate particles appear?

[qh_Lu]

Hi, everyone!
In order to improve the computational efficiency of cryoSPARC, bin particles are often used in 2D classification.
In my case, I re-extracted the particles selected after 2D classification in order to eliminate bins. It was found that the newly extracted particles contained a large proportion of duplicate particles, making the FSC curve unable to return to zero at high frequencies.
My doubt is: Duplicate particles can be eliminated in 2D classification, why do the newly extracted 2D classified particles contain a large proportion of duplicate particles?

Looking forward to your reply, thank you!!

[olibclarke]

Try to remove the duplicates using the separate remove duplicates job prior to extraction. I don’t think the on-the-fly duplicate removal in Class2D currently works for binned particles.

[qh_Lu]

Thank you for your answer.
Could you please tell me why you think that duplicate removal in 2D classification is not suitable for binned particles?

[olibclarke]

Remove duplicates in Class 2D (query)

Hi @olibclarke, I think we may have found out the cause of this discrepancy. Based on the pixel sizes in the 2D class event log you attached, it looks like downsampling was likely used. Unfortunately the implementation of remove duplicates in 2D Classification (in the case of downsampled data) uses an incorrect pixel size when converting particle coordinates into physical angstrom distances, and this would cause it to reject way fewer particles than expected. We’ll aim to fix this in an upcomin…

[qh_Lu]

You successfully let me understand the reason of this problem, thank you very much! !