EPISCORE

EPISCORE performs virtual microdissection of bulk tissue DNA methylomes by leveraging single-cell RNA-Seq atlases and a probabilistic epigenetic model of gene regulation to infer cell-type-specific methylation signals and estimate cell-type proportions in solid tissues.

Key Features:

• R package implementation: Implemented as an R package for computational analysis of methylation data.
• Single-cell RNA-Seq-derived reference matrix: Generates tissue-specific DNA methylation reference matrices from single-cell RNA-Seq tissue atlases.
• Probabilistic epigenetic model: Uses a probabilistic epigenetic model of gene regulation to relate expression-derived signatures to methylation patterns.
• Deconvolution and proportion estimation: Performs deconvolution to quantify proportions of different cell types within bulk tissue DNA methylomes.
• Cell-type-specific differential methylation detection: Identifies cell-type-specific differential methylation signals from bulk methylation profiles.
• No single-cell methylome required: Circumvents the need for direct single-cell DNA methylome data by integrating single-cell RNA-Seq with bulk methylation profiles.
• Single-cell-type resolution for solid tissues: Enables virtual microdissection at single-cell-type resolution for any solid tissue.
• Cross-study validation: Validated across multiple epigenome studies and diverse tissue types.

Scientific Applications:

• Bulk methylome deconvolution: Deconvoluting bulk tissue DNA methylomes to resolve cell-type composition.
• Cell-type proportion estimation: Estimating proportions of constituent cell types from bulk methylation data.
• Detection of cell-type-specific differential methylation: Detecting differential methylation signals attributable to specific cell types within heterogeneous tissues.
• Interpretation of epigenetic regulation: Facilitating interpretation of epigenetic regulation across different cell types and tissues.
• Support for biological and clinical studies: Supporting investigations of regulatory mechanisms relevant to basic biology and potential clinical research applications.

Methodology:

Generates tissue-specific DNA methylation reference matrices from single-cell RNA-Seq tissue atlases, integrates these matrices with bulk tissue DNA methylation profiles, and applies a probabilistic epigenetic model of gene regulation to perform deconvolution and infer cell-type proportions and cell-type-specific differential methylation.