search

3dem / relion

Image-processing software for cryo-electron microscopy
https://relion.readthedocs.io/en/latest/
GNU General Public License v2.0
456 stars 203 forks source link

relion_helix_inimodel2d #1210

Open sprombhu opened 4 days ago

sprombhu commented 4 days ago

Hello,

I ran this

relion_helix_inimodel2d --i IniModel/run1_it005.star \ --iniref 1@IniModel/run1_it005_reconstructed.mrcs --angpix 1.15 \ --maxres 6 --search_angle 2 --step_angle 0.5 --mask_diameter 250 \ --j 6 --iter 5 --o IniModel/run2

and got this error

ERROR: Projector::computeFourierTransformMap%%ERROR: Dimension of the data array should be 2 or 3.

I went to see possible solutions, and found this

git clone heep://bforsbe@bitbucket.org/scheres/relion-3.0_beta.git

might help but instead I have this

Cloning into 'relion-3.0_beta'... fatal: Unable to find remote helper for 'heep'.

Would I please get some helps here?

Kind regards, -S

biochem-fan commented 4 days ago

relion-3.0_beta

This is an extremely old version which we no longer support. Using this version probably does not solve your problem.

sprombhu commented 4 days ago

I am actually working on 4.0. Will the command work the same way?

sprombhu commented 4 days ago

Originally was because of this

ERROR: Projector::computeFourierTransformMap%%ERROR: Dimension of the data array should be 2 or 3

from this

relion_helix_inimodel2d --i IniModel/run1_it005.star --iniref 1@IniModel/run1_it005_reconstructed.mrcs --angpix 1.15 --maxres 6 --search_angle 2 --step_angle 0.5 --mask_diameter 250 --j 6 --iter 5 --o IniModel/run2.

biochem-fan commented 4 days ago

It should be but I am not an expert on helical processing.

Did you read https://pmc.ncbi.nlm.nih.gov/articles/PMC7008511/?

sprombhu commented 4 days ago

Thank you very much for your replies. Instructions and commands I use were given from that paper.

biochem-fan commented 4 days ago

I guess you are using --iniref in a wrong way.

The paper says:

Optionally, the user can provide an initial 2D reference image using the --iniref option. Provided that it is rescaled to the same pixel size and re-windowed to the same box size as the input 2D class averages (using the relion_image_handler program), this initial reference could, for example, come from the low-resolution XY structure obtained with the 2D class average spanning an entire cross-over.

Isn't IniModel/run1_it005_reconstructed.mrcs a 3D volume?

sprombhu commented 4 days ago

That is totally why. Thank you!

scheres commented 1 day ago

No, iniref should be a 2D image (we're trying to reconstruct the xy-crosssection. I'm not sure the n@stack.mrcs syntax works though, could you make an individual image first? relion_image_handler --i 1@IniModel/run1_it005_reconstructed.mrcs --o iniref.mrc