MPFE

MPFE estimates the distribution of DNA methylation patterns from bisulphite sequencing count tables to quantify cytosine methylation states at genomic loci while accounting for bisulphite non-conversion and sequencing read errors.

Key Features:

• Pattern estimation from counts: Estimates the distribution of methylation patterns from a table of counts derived from bisulphite sequencing reads.
• Non-conversion and error modelling: Explicitly models bisulphite non-conversion rates and sequencing read error rates to separate true methylation signals from artefacts.
• Read-level state analysis: Analyzes sequences of methylation states along reads rather than relying solely on average methylation at individual positions.
• Statistical modelling: Employs a statistical model developed for accurate quantification of methylation pattern distributions.
• Locus-level inference: Provides estimates of the underlying distribution of methylation patterns at individual loci.
• Implementation: Implemented as an R Bioconductor package.

Scientific Applications:

• Epigenetics studies: Quantifying diversity and distribution of DNA methylation patterns in epigenetic analyses.
• Gene regulation research: Investigating associations between methylation pattern heterogeneity and gene regulation.
• Methylation landscape characterization: Characterizing the complexity of methylation landscapes across biological samples at the read and locus level.

Methodology:

Fits a statistical model to bisulphite sequencing count tables that analyzes methylation-state sequences along reads and explicitly models bisulphite non-conversion rates and sequencing read error rates; implemented in R Bioconductor.