MATCHER: manifold alignment reveals correspondence between single cell transcriptome and epigenome dynamics.

Area of Science:

• Genomics
• Epigenetics
• Computational Biology

Background:

• Single cell technologies reveal cellular heterogeneity in transcriptomics and epigenetics.
• The relationship between these molecular layers at the single-cell level remains largely undefined.
• Integrating multi-omic data from individual cells is crucial for understanding cellular function.

Purpose of the Study:

• To present MATCHER, a novel computational approach for integrating diverse single-cell measurements.
• To infer multi-omic profiles by aligning transcriptomic and epigenetic data from distinct cells.
• To investigate the interplay between the epigenome and transcriptome in stem cells.

Main Methods:

• Developed MATCHER, a manifold alignment technique for single-cell data integration.
• Applied MATCHER to single-cell transcriptomic (scM&T-seq) and epigenomic (sc-GEM) datasets.
• Validated MATCHER's ability to predict correlations without cell-to-cell correspondence.

Main Results:

• MATCHER accurately inferred single-cell multi-omic profiles.
• Confirmed true single-cell correlations between DNA methylation and gene expression.
• Identified novel insights into the dynamic interplay between epigenome and transcriptome.
• Demonstrated efficacy on embryonic stem cells and induced pluripotent stem cells.

Conclusions:

• MATCHER provides a robust framework for integrating single-cell transcriptomic and epigenomic data.
• The approach accurately predicts molecular correlations, advancing multi-omic single-cell analysis.
• MATCHER facilitates deeper understanding of epigenetic regulation of gene expression in cellular development and reprogramming.