Chromatin-accessibility estimation from single-cell ATAC-seq data with scOpen.

Area of Science:

• Genomics
• Computational Biology
• Epigenetics

Background:

• Single-cell ATAC-seq (scATAC-seq) is crucial for studying cell-specific chromatin accessibility.
• Data sparsity, caused by DNA loss during protocols, is a major limitation of scATAC-seq.
• This sparsity hinders accurate downstream analysis and biological interpretation.

Purpose of the Study:

• To develop a computational method, scOpen, to address sparsity in scATAC-seq data.
• To improve the imputation and quantification of open chromatin regions in sparse datasets.
• To enhance downstream analyses including clustering, visualization, and regulatory element identification.

Main Methods:

• scOpen utilizes regularized non-negative matrix factorization.
• The method imputes and quantifies open chromatin status from sparse scATAC-seq data.
• The performance of scOpen was evaluated on its impact on downstream analyses.

Main Results:

• scOpen effectively imputes missing data in sparse scATAC-seq experiments.
• The method significantly improves clustering, visualization, and identification of regulatory features.
• Application to kidney fibrosis revealed the role of Runx1 in fibroblast to myofibroblast differentiation.

Conclusions:

• scOpen is a powerful computational tool for overcoming sparsity in scATAC-seq data.
• The method enhances the accuracy and biological insights derived from scATAC-seq.
• scOpen facilitates the dissection of regulatory mechanisms in diseases like kidney fibrosis, identifying key players such as Runx1.