After merging batches, I got very segmented umap in scanpy

Question

[MohammadAlkadi]

I merged four samples of 10X genomics and I did batch correction using harmony which converged after 4 iterations. I get this very segmented umap. But If I do each sample separately I get a normal clustering with (~28 clusters). What can be the reason?

Thank you in advance.

Answer 1

[MohammadAlkadi]

Very sorry for the inconvenience. It turned out I merged the same batch 4 times.