Unsupervised clustering and epigenetic classification of single cells.

Area of Science:

• Genomics
• Epigenetics
• Computational Biology

Background:

• Cellular-level epigenetic heterogeneity is crucial in genomics.
• Single-cell ATAC-seq (scATAC-seq) reveals epigenetic variability but faces technical challenges.
• Distinguishing cell types in heterogeneous populations requires robust analytical methods.

Purpose of the Study:

• To introduce scABC, an R package for unsupervised clustering of single-cell epigenetic data.
• To enable accurate classification of scATAC-seq data.
• To discover cell-identity-specific open chromatin regions.

Main Methods:

• Development of scABC, an R package utilizing unsupervised clustering algorithms.
• Application of scABC to single-cell ATAC-seq datasets.
• Analysis of open chromatin patterns for cell type identification.

Main Results:

• scABC effectively clusters single-cell epigenetic data.
• The package facilitates clear classification of cells within heterogeneous populations.
• scABC aids in the discovery of cell-specific regulatory elements.

Conclusions:

• scABC provides a robust solution for analyzing scATAC-seq data.
• The package improves cell type classification and identification of cell-identity-specific genomic regions.
• scABC advances the study of epigenetic heterogeneity at the single-cell level.