dcjones
Clarity for Visium HD+ cellpose-sam
Hi @dcjones , thank you for finally adding functionality for Visium HD! Grateful for all your hard work.
It's a bit unclear how to use cellpose-sam output with Visium HD. In your cellpose xenium script, you are doing some tiling etc, which I'm not sure if I need for my data.
For me, I have been successfully able to run cellpose-sam independently, and generate the .npy files for the masks, and the component 2 of the flows, for probability. Can it not directly be used with the Visium HD output, given that the exact same image was used in segmentation, and provided to spaceranger (the spaceranger segmentation was unsatisfactory).
If you could please highlight the steps for using masks and cell probability generated independently, into PROSEG, that would be really helpful!
Thank you and warm regards
Jairaj Mathur
I've just added a little example of how to do this here: https://github.com/dcjones/proseg/blob/main/examples/visiumhd-cellpose.sm
