Preprocessing of adata.X for embedding generation with CancerFoundation

[lakii3741]

Dear authors,

Thank you for developing this great model! I am currently working on generating embeddings for my scRNA-seq data using the CancerFoundation model, as outlined in tutorial/embeddings_tutorial.ipynb. My aim is to understand the recommended preprocessing steps for the AnnData object, in particular the adata.X matrix, before it is passed to the embed function (e.g. filtering out genes expressed in less than 10 cells).
I have looked through both

1. the tutorial/embeddings_tutorial.ipynb tutorial
2. the CancerFoundation paper here

but could not find any instructions or recommendations for it.
Thank you for your guidance in advance!

Best regards,
Justina