Optimizing reconstruction parameters
Forums
Hi everyone,
Could you please share some of your experience on how do you pick the best reconstruction parameters for a given map, notably, thresh_reconst and PBC?
Although I try to optimize these parameters following some kind of systematic search while assessing the map visually and by the FSC, I am concerned whether my reconstructions are the best they can be. I'd like to know how do people pick these values and whether there is any "optimal" way of setting them.
Thanks a lot.
Trying different values as
Trying different values as you have done is also what we do. The FSC should be a good indicator of map quality and which parameters are the best. You should also look at the final maps and verify that a better FSC corresponds with a better map.
As a note on the side, in most cases the improvements in the map by optimizing the PBC and score threshold parameters are not very significant. Therefore, the precise values are probably not important.
PBC optimization
In reply to Trying different values as by niko
I have, in the past with a small particle that is quite heterogeneous, gotten significant improvements from PBC optimization (from ~5.5 to ~5 Angstroms). Most often though, it only results in a few Fourier pixels difference in resolution, at best.
HTH,
Axel