Figure 1.

Workflow: ($\mathbf{a}$) The scATAC-seq data serve as the input. ($\mathbf{b}$) The data are partitioned into GAGAM contributions, resulting in promoter and enhancer matrices. ($\mathbf{c}$) For each contribution, utilizing the JASPAR database, both motif enrichment and TF footprint scores are determined for all TFs expressed in the dataset. ($\mathbf{d}$) Cells are aggregated based on cell-type annotations. ($\mathbf{e}$) The two matrices obtained for each contribution are then compared to the expression matrix to analyze their correlation. ($\mathbf{f}$) Results analysis reveals specific correlations and dynamics between TF expression and its motif information, with plots showing the TF motif (A), the motif enrichment (C), and the footprint scores (B,D).