Depth-corrected multi-factor dissection of chromatin accessibility for scATAC-seq data with PACS

Area of Science:

• Genomics
• Computational Biology
• Biostatistics

Background:

• Single-cell ATAC-seq (scATAC-seq) experiments are becoming more complex, involving multiple factors like genotype and cell type.
• Existing methods struggle with the statistical challenges of sparse scATAC-seq data and complex experimental designs.

Purpose of the Study:

• To introduce a novel zero-adjusted statistical model, Probability model of Accessible Chromatin of Single cells (PACS).
• To enable complex hypothesis testing for factors influencing chromatin accessibility in sparse scATAC-seq data.
• To improve statistical power and control false discovery rates in differential accessibility analysis.

Main Methods:

• Development of a zero-adjusted statistical model (PACS).
• Implementation of complex hypothesis testing for accessibility-modulating factors.
• Accounting for data sparsity and variations in single-cell sequence capture.

Main Results:

• PACS demonstrates superior performance in differential accessibility analysis, with 17% to 122% higher power than existing tools.
• The model effectively controls the false positive rate.
• PACS successfully performs supervised cell type annotation, compound hypothesis testing, batch effect correction, and spatiotemporal modeling.

Conclusions:

• PACS provides a robust framework for analyzing complex scATAC-seq data.
• The model reveals previously undiscovered biological insights from diverse tissue datasets.
• PACS enhances the statistical rigor and analytical power for scATAC-seq studies.