Leginon

1. Trivial: When finishing an imaging target list, especially when using queue timeout, have the stage reset to 0,0 in x,y in addition to returning to 0 deg tilt.

2. As an alternative to the ‘crosses’ generated by Final_Raster_Targeting for imaging targets, have a ‘box’ superimposed on the image instead to indicate the area covered by the camera for each target. This could be an alternative to the crosses or displayed in addition to the targets.

3. Add the ‘Fix Exposure Time’ node to Section3Step. It was in one of the screenshots of another application Anchi showed during the one-day workshop on multiple grid. In talking to Anchi about it, it sounds exactly like something we were interested in during last year’s contract with you guys What statistics? Max and mean, per section by section thickness. May be on the plastic.

4. Pre-imaging baking of the sections. For stability during imaging on slot grids, we typically do a little pre-burning of the sections prior to start of the imaging run. At ‘ex’, I spread the beam to cover the entire slot, then move the sections into the beam for several minutes, then move the next set of sections, etc until I’ve covered all of them. If this process could be automated using the Grid Atlas (or Tissue Centering) targets by having the stage ‘wander’ over the sections, that would be very helpful do this after each tilt/rotation? Or just the beginning?

5. A test was put in to see if the camera was inserted before imaging, and insert it if the operator forgot to. A similar test if the screen is raised before imaging, and do so if the operator forgot to raise it.

6. Raster targeting is now ellipse based, which is ideal for our fly brain sections since the tissue is generally oval in shape. This has reduced the number of images by 50% over the previous method of setting up a square/rectangular raster pattern. HOWEVER, other projects here that use Leginon would actually benefit for targeting purposes if there were a square/rectangular pattern option. Can this be ‘restored’ to Section3Step as an option?

7. Occassionally due to issues with the CompuStage (a stage tracking error), targets in a raster pattern are skipped and not imaged. Leginon generated messages in the Final_Raster log window about this. We would like Leginon to send an email on a stage tracking error resulting in target(s) not being imaged. An email reporting enhancement was added to Leginon during last year’s development contract for images in which the mean was greater or less than a specified range – indicating a broken film or screen not raised during imaging. What will you do with the e-mail? Are you going to come in and pause or repeat or ….

8. Add ellipse-type tool to Tissue_Centering to determine centroid of ROI for subsequent Final_Raster_Target node. This would take some of the extra imaging out of the final imaging process.

9. For our multiple tilt imaging, have Leginon return the stage to 0 tilt, and x,y of 0,0 after last target is submitted in Tissue_Centering or upon submission of queue from Tissue_Centering. Why? Once submitted it is going to move away immediately.

10. Ability to remove specific targets from patterns generated by Final_Raster_Targeting. The ellipse tool works great, but there are still instances that to cover all the tissue, 5 or more unnecessary targets are generated over empty plastic. Spread over 6000 sections, this is ~30,000 images that don’t have to be taken. These “extra” targets could easily be removed via a right click during user acceptance of each raster pattern.