error related size

[Sijaz1]

First of all, thank you for developing and sharing this excellent pipeline. It works very well with the example datasets.

I am currently trying to use it with my own large mouse atlas dataset (in .h5ad format), but I am encountering an issue related to the file size. When I use your example files, the pipeline completes successfully within about one hour.

However, when I use my dataset, the pipeline keeps running for 4 to 5 days without producing results, especially during the final evaluation step. It seems to get stuck there.

Could you please advise if there are any size limitations or recommended settings for handling large .h5ad files? Or if there are any specific parameters I should adjust to make it run efficiently?

Thank you again for your great work and support.