Templates from alpha fold predictions

[DarioSB]

Indeed, I have experimented a lot with 2D classification parameter to get better classes, which worked, but one of my goals was to get other views then the biased orientation that I have at the moment.
Thanks for the advice on ab-initio

[miwoodso]

I just want to mention that your model needs to be in the center of the “map of the right size” you are using for onGrid #X. If the two aren’t centered, your density can be cut off, or you might end up with an empty volume and blank templates.

[J6H55]

also encounter this problem. is there any suitable command to form a suitbale map in chimeraX?
my error is ValueError: could not broadcast input array from shape (69,39,35) into shape (34,34,34)

[olibclarke]

get the daily version. There is now a “cube” argument for molmap which will ensure the map is a cube (which is the issue here)

[iram]

How to change non cube to cube?

[rbs_sci]

I’d usually pad the box out with relion_image_handler --new_box [newBoxSize] --i [input map] --o [output map] and then re-center and crop back if needed.

[olibclarke]

or just resample in Chimera on a cubic map (any EM map)

[rbs_sci]

For some reason I rarely use Chimera for that. I got into the habit of doing it in the command line with first EMAN2 then later RELION. Big boxes in Chimera over SSH is torture, that’s probably why.