Normalizing data based on spike-in cells


I have several separate 10X datasets that each have a small population (~1,000 cells) of 293T spike-in cells that have a higher copy number compared to my sample cells.

I am able to isolate them from my data by clustering and identify them using InferCNVpy, but I am wondering how I might use this spike-in population to normalize my sample cells in the same dataset.

The idea is to be able to compare cell types in my sample cells across discrete datasets that all contain the same spike-in cell type.

Please let me know if you can help provide insight into how I might use Scanpy to do this.

Thank you