Preferred method for generating initial Eulers

Submitted by sstagg on Tue, 2015-05-19 07:02

Forums

Frealign

Hi all,

I'm wondering what your preferences are (i.e. are there best practices) for generating initial Eulers. I've been using Frealign for years, but I've often wondered how much initial Eulers affect the final reconstruction. Do y'all boot off of an external program or use MODE 3 or 4?

Thanks,
Scott

Axel Tue, 2015-05-19 12:01

Hi Scott,

I used to use IMAGIC projection matching for my ribosomes, but found that I got similar if not better results using Frealign's mode 4 (unpublished data).
If your molecule is symmetric and fairly large, you can excellent results by just using Mode 4 (and it may in fact be the fastest way to assign initial Eulers).

Recently, I have been running a few rounds of Relion processing, prior to exporting to Frealign, though I haven't run a good A/B test to determine which is best, Frealign Mode 4, or Relion with binned data.

Best,

Axel

sstagg Tue, 2015-05-19 17:42

Thanks, Axel. When you use Mode 4, do you just use the defaults or do you change some of the search parameters?

Axel Tue, 2015-05-19 18:36

The defaults are very similar to what I found worked best for the ribosome data I had collected, and should be a good start. If you find you are having trouble, there are a few lines in the README that explain what parameters are good to play with (specifically Res refine, Res Reconstruct, Bfactor, Amplitude contrast).