Additional file 1 of scCompressSA: dual-channel self-attention based deep autoencoder model for single-cell clustering by compressing gene–gene interactions

Additional file 1. Evaluation metrics of replication experiments obtained by the scCompressSA and SOTA single-cell clustering approaches, including Seurat, SC3, scCAEs. Ten groups of benchmark scRNA-seq datasets with cell type labels have been employed in model comparison. Among these datasets, multiple PBMCs data from human were investigated in this study.