Fig. 4: QCM/QCC plots using different normalizations for the SCA input counts table.
From: Sparsely-connected autoencoder (SCA) for single cell RNAseq data mining
a Log10 transformed, (b) Centered log-ratio normalization (CLR), (c) relative log-expression (RLE), (d) full-quantile normalization (FQ), (e) sum scaling normalization (SUM), (f) weighted trimmed mean of M-values (TMM).
