DXM

DXM deconvolves DNA methylation profiles from bisulfite sequencing data to resolve major allelic subpopulations in heterogeneous samples for studying epigenetic heterogeneity such as cancer clonal expansion and disease progression.

Key Features:

• Local Deconvolution: Performs local deconvolution of DNA methylation data for specific genomic regions without requiring prior knowledge of the number or types of subpopulations.
• No Prior Knowledge Required: Does not require prior information about expected cell types or the count of subpopulations within a heterogeneous sample.
• Benchmarked Performance: Has been benchmarked against other methodologies, demonstrating superior performance in identifying allelic subpopulation methylation profiles.
• Experimental Validation: Predictions were experimentally validated in four Diffuse Large B-Cell Lymphomas (DLBCLs).
• Proof-of-Concept Application: Applied to a cohort of 31 DLBCLs, establishing a relationship between allelic subpopulation methylation profiles and disease relapse.

Scientific Applications:

• Cancer epigenetics: Resolve epigenetic heterogeneity in heterogeneous tumor samples such as DLBCL to study clonal evolution and disease relapse.
• Subpopulation-specific epigenetic analysis: Identify subpopulation-specific DNA methylation changes that may drive disease progression or influence treatment outcomes, including therapeutic resistance.

Methodology:

DXM performs local deconvolution on bisulfite sequencing data to infer methylation profiles of major allelic subpopulations without requiring prior knowledge of subpopulation number or cell types.