scCASE: Accurate and interpretable enhancement for single-cell chromatin accessibility sequencing data

Single-cell chromatin accessibility sequencing (scCAS) has emerged as a valuable tool for interrogating and elucidating epigenomic heterogeneity and gene regulation. However, scCAS data inherently suffers from limitations such as high sparsity and dimensionality, which pose significant challenges for downstream analyses. Although several methods have been proposed to enhance scCAS data, there are still challenges and limitations that hinder the effectiveness of these methods. Here, we propose scCASE, a scCAS data enhancement method based on non-negative matrix factorization. scCASE incorporates an iteratively updating cell-to-cell similarity matrix to enhance scCAS data. Through comprehensive experiments on simulated and multiple real datasets, we demonstrated the advantages of scCASE over existing methods for scCAS data enhancement. The scCAS data enhanced by scCASE can be effectively utilized for downstream analyses, such as cell clustering and data visualization. Besides, cooperating with extensive function enrichment, motif enrichment, and heritability enrichment, we showed that the interpretable cell type-specific peaks identified by scCASE can provide valuable biological insights into cell subpopulations. Moreover, to fully leverage the large compendia of available omics data as a reference, we further expanded the scCASE method to scCASER, which enables the incorporation of external reference data to improve enhancement performance. We also introduced various approaches to construct reference data, facilitating the broad application of our method in enhancing scCAS data.