search

IsoNet-cryoET / IsoNet

Self-supervised learning for isotropic cryoET reconstruction
https://www.nature.com/articles/s41467-022-33957-8
MIT License
67 stars 12 forks source link

Pixel size of 0 in corrected tomograms #14

Open mvorlander opened 3 years ago

mvorlander commented 3 years ago

Hi there,

I have an issue with the predict step of isonet. I am running it on the same tomograms that I used for training, the pixel size is 10.00 Å/px and this information is also present in the star file that I submit during the predict step. However, the corrected_tomograms have a pixel size of 0 and when I open them they are accordingly on the incorrect scale compared to unprocessed tomograms.

Is anyone else having this problem and suggestions what causes this?

procyontao commented 3 years ago

Hi,

This is the problem of the program. The pixel size was omitted during predicting. This should be fixed soon.

Alternately, if you can open the tomogram with chimera(x), use 'volume voxelSize' command can change the pixel size.

cheers

mvorlander commented 3 years ago

Ok, thanks for the reply! :)