Auto-picking parameter recommendation for the class dataset (reference free version) : Penn EM Class

Auto-picking parameter recommendation for the class dataset (reference free version) Print

Created by: hee jong kim

Modified on: Sun, 14 Mar, 2021 at 9:45 PM

As you may tried multiple round of manual picking and auto picking along with subsequent stages (e.g. 2D/3D classifications), I hope you realized that the importance of getting the particle picking right for the rest of data analysis.

Very much credit should go to the former TA, Trevor van Eeuwen, for the parameters.

This should be a reasonable start and please feel free to test out different combinations from this setting to get a better one.

NOTE: Please make sure to follow every single parameter value in I/O, Laplacian, and autopicking tabs.

In I/O tab, provide your ctf estimation result.

In Laplacian tab, set the parameters shown below.

SKIP References tab as this FAQ page is for reference free picking.

In autopicking tab, set parameters as follows.

That's It.

Now submit your job with a single cpu node. (48 MPI with multi_cpu.sh)

Here's an example result from this parameter.

Hope you have a fun with it and improve furthermore.

Hee is the author of this solution article.

Did you find it helpful? Yes No

Auto-picking parameter recommendation for the class dataset (reference free version) Print

Related Articles