Matching queried single-cell open-chromatin profiles to large pools of single-cell transcriptomes and epigenomes for...

Area of Science:

• Genomics
• Computational Biology
• Cell Biology

Background:

• Large-scale single-cell data analysis requires advanced tools for cell annotation.
• Matching single-cell epigenome profiles to reference datasets is a significant computational challenge.

Purpose of the Study:

• To develop and validate scEpiSearch, a novel computational tool for searching, comparing, and classifying single-cell open-chromatin profiles.
• To demonstrate scEpiSearch's utility in identifying cellular heterogeneity and regulatory states across diverse biological contexts.

Main Methods:

• Development of scEpiSearch for querying single-cell open-chromatin data against large reference datasets.
• Benchmarking scEpiSearch's performance against existing methods for accuracy and coembedding.
• Application of scEpiSearch to analyze epigenome profiles from K562 cells, acute leukemia patient samples, and embryonic stem cells (ESCs).

Main Results:

• scEpiSearch demonstrated superior accuracy in searching and coembedding single-cell profiles compared to other methods.
• The tool successfully identified heterogeneity, multipotency, and dedifferentiation in K562 cells and leukemia samples.
• scEpiSearch revealed specific ESC subpopulations poised for endoplasmic reticulum stress and unfolded protein response.

Conclusions:

• scEpiSearch effectively addresses the challenge of integrating large single-cell epigenome datasets.
• The tool enables the identification and study of cell-specific regulatory states using epigenomic information.
• scEpiSearch provides a powerful platform for exploring cellular heterogeneity and function at single-cell resolution.