Skip to main content
Fig. 1 | Genome Biology

Fig. 1

From: scBFA: modeling detection patterns to mitigate technical noise in large-scale single-cell genomics data

Fig. 1

Single-cell binary factor analysis (scBFA) outperforms quantification models. Performance is measured via cross-validation of cell type classifiers trained on scRNA-seq benchmark data in the respective embedding spaces of each method, as a function of the number of latent dimensions specified. scBFA is a top performer in 13 out of 14 datasets. Datasets from left to right, top to bottom: Dendritic, Pancreatic, DC, mESCs, HSPCs, MGE, Intestinal, MEM-T, H7-ESC, LSK, Myeloid, HSCs, PBMC, and LPS (see Additional file 1: Table S1 for a detailed description of benchmarks)

Back to article page