Saving a Relion5 particle list shifts all particles

[ASKrebs]

Hello,

we are trying to pick points along a line in the GUI of ArtiaX. It generates the line properly, we can create particles that look okay and then we save the list as a Relion5 star file.
However, it doesn't seem to save the particles at the correct coordinates. So, if we close the list and reopen, it shows the particles somewhere very far away from the tomogram. We also haven't figured out with what factor the coordinates are off - it is not by the binning or the pixel size or the shift to the center of the tomogram.
If we do the same workflow but instead save as .em or .coords, it saves correctly and reopening also shows them in the correct position as well.

Does someone else have this problem?

[Phaips]

I have encountered this before but not always. Sometimes saving the particles.star works fine but sometimes the coordinates are way off if you check the .star files. I haven't yet identified the part of the code which causes this problem. I think it depends if you edit an opened .star file from relion5 or if you generate one from scratch.

one fix is to run for example:
save path/to/particles.star partlist #1.2.1 voldim x,y,z voxelsize 1.000 prefix None suffix None tomonumber 1 prior false

with voxelsize 1 and your voldim is your tomogram dimension in binned pixels times your pixelsize
for example 1024,1024,512 (bin4) with pixelsize 9.68 would become voldim 9912.32,9912.32,4956.16 voxelsize 1.000

hope this helps