Doug Shepherd

> This is more or less the way we do it with the software running on the iSIM atm. Define the DAQ data/task for the full sequence and start on...

> Did you start experiencing this with a certain version of multiview-stitcher? This has always been an issue with multiview-stitcher, it just took us a long time to narrow it...

Sorry for dropping this, I had some travel, got sick, and then the semester started. Thanks for your work! We will test it out on some datasets and get back...

For one channel, things definitely seem improved. For multiplexed imaging (e.g. 18 channels of iterative labeling), the memory still runs away and then the kernel dies. Is there a way...

> Thanks for the feedback. Interesting, so if I understand you correctly, things work well with a single channel but fail for many channels. Correct. I believe the individual FOVs...

Saving as independent channels works if we downsample the fused image output spacing. Strangely, if we use the the full resolution data it saves all of the chunks, but everything...

Thanks! We segment cells (poyDT marker), not nuclei, for our initial assignment. It looks like we should use the `--use-cell-initialization` flag? We have a high-memory server, so we can give...

We are coming back around to trying this with the newest updates you've made. We have a 16-plex smFISH experiment in cell culture we performed using our 3D approach. The...

Ah - I think we figured it out. There were some erroneous `cell_id` entries for transcripts outside of cells. So that was probably throwing off the algorithm.

@grlee77 - I had time yesterday to test the proposed fix using admin privileges. I was able to get the install to work after copying `where.exe` to `which.exe` so that...