What types of clustering is required?

[JiahuaQu]

Hi, we are considering using the SoupX, but we are confused about the clusters requirement. Because we usually analyze single-cell data that are generated by 10X in Seurat, we didn't do the downstream analysis in CellRanger, such as clustering. We will cluster cell types in Seurat. In this case, I don't think there's any cluster information from 10X folder. However, the data should have already been processed, filtered and normalized in Seurat before clustering, so I am confused what types of count matrix (raw, normalized, scaled) and what types of clustering are required by the SoupX?
Is there any suggestion how to connect SoupX with Seurat? Thank you.