scDC_noClustering() or scDC_Clustering? what is the difference?

[docjim86]

Hi,
I am using scDC to do the DA analysis, and i am a little confused about "scDC_noClustering" and "scDC_Clustering" cause i can not find a description indicating the difference between these two Function.
My sc-data was analysis by the Seurat toolkit and annotated into 14 celltype. Is that means I already got 14 "Clusters"? So my DA pipline should be scDC_Clustering followed by fitGLM ? Is it wrong i use scDC_noClustering ?
Thank you for the scDC package. Appreciate for your reply.

[docjim86]

Hi, I am using scDC to do the DA analysis, and i am a little confused about "scDC_noClustering" and "scDC_Clustering" cause i can not find a description indicating the difference between these two Function. My sc-data was analysis by the Seurat toolkit and annotated into 14 celltype. Is that means I already got 14 "Clusters"? So my DA pipline should be scDC_Clustering followed by fitGLM ? Is it wrong i use scDC_noClustering ? Thank you for the scDC package. Appreciate for your reply.

sorry, I found that

performed without clustering,the estimation does not take into account for the variability from clustering algorithm, and only takes variability from the population into account. When performed with clustering, the estimation takes both into consideration.