batch effect correction and harmization #409
Description
Thanks for the nice tool you provided! I have a question regarding my dataset! I have a scRNA-seq dataset from 40 human BM samples with around 300k cells, I am interested to use another clustering method other than Seurat which I used previously and came across Celda! My question is does celda uses any algorithm to correct the batch effect and integrate/harmonized the data? my data is from 6 different batches so I needed to correct the batch effect and integrated and used the harmony through Seurat! so I guess my questions are:
1- If I use Celda using only a merged seurat object (not integrated, not batch corrected) how does celda take care of this?
2- Can I use the singleCellExperiment Object from integrated and corrected Seurat object and does this really matter? cause I know Celda will use its own algorithm for clustering and I am not sure if the corrected and integrated object really matters here since I guess celda will use the count data, yes?
I appreciate your input.
best,
Sogand