Skip to main content
Fig. 2 | Genome Biology

Fig. 2

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

Fig. 2

Removing batch effects in simulated scRNA-seq data. a UMAP visualizations of results for Experiment removal2 on 2D Gaussian dataset, where Type1 from Batch1 and Type4 from Batch2 were removed. BERMUDA_0.85 and BERMUDA_0.90 represent results of BERMUDA with Sthr = 0.85 and 0.90, respectively. b UMAP visualizations of results for Experiment removal2 on Splatter dataset, where Type1 from Batch1 and Type 4 from Batch2 were removed

Back to article page