Best practices for single-cell histone modification analysis

Background Single-cell histone post translation modification (scHPTM) assays such as scCUT&Tag or scChIP-seq allow single-cell mapping of diverse epigenomic landscapes within complex tissues, and are likely to unlock our understanding of various epigenetic mechanisms involved in development or diseases. Running an scHTPM experiment and analyzing the data produced remains, however, a challenging task since few consensus guidelines exist currently regarding good practices for experimental design and data analysis pipelines. Methods We perform a computational benchmark to assess the impact of experimental parameters and of the data analysis pipeline on the ability of the cell representation produced to recapitulate known biological similarities. We run more than ten thousands experiments to systematically study the impact of coverage and number of cells, of the count matrix construction method, of feature selection and normalization, and of the dimension reduction algorithm used. Results The analysis of the benchmark results allows us to identify key experimental parameters and computational choices to obtain a good representation of single-cell HPTM data. We show in particular that the count matrix construction step has a strong influence on the quality of the representation, and that using fixed-size bin counts outperforms annotation-based binning; that dimension reduction methods based on latent semantic indexing outperform others; and that feature selection is detrimental, while keeping only high-quality cells has little influence on the final representation as long as enough cells are analyzed.