Skip to main content
Fig. 4 | Genome Biology

Fig. 4

From: BERMUDA: a novel deep transfer learning method for single-cell RNA sequencing batch correction reveals hidden high-resolution cellular subtypes

Fig. 4

Removing batch effects in scRNA-seq data of PBMCs. a Expression patterns of marker genes of immune cells colored by log-transformed TPM values. b The UMAP visualization of results produced by BERMUDA colored by batches. Different cell types were identified by analyzing expression patterns of maker genes and are highlighted by dashed circles. BERMUDA correctly combined the corresponding cell types between different batches while preserved cell types not shared by both batches as separate clusters. c The UMAP visualization of pan T cell batch. No obvious clustering structure was observed when visualizing pan T cell batch individually

Back to article page