Skip to main content
Fig. 4 | BMC Bioinformatics

Fig. 4

From: Automated annotation of rare-cell types from single-cell RNA-sequencing data through synthetic oversampling

Fig. 4

Figure showing a comparison between the distribution of synthetic cells (purple) generated by sc-SynO and original input cells (blue) for model training. The LoRAS algorithm creates synthetic cells by generating convex combinations of Gaussian noise added to rare-cell samples. This can be thought of as a random weighted average of multiple cells. In this figure, we highlight the generated synthetic and real cells to enable a visual comparison

Back to article page

BMC Bioinformatics

ISSN: 1471-2105

Contact us