Identifying Differential Methylation in Cancer Epigenetics via a Bayesian Functional Regression Model

Area of Science:

• Genomics
• Epigenetics
• Bioinformatics

Background:

• DNA methylation is vital for gene regulation, disease susceptibility, and treatment outcomes.
• Next-generation sequencing provides single-nucleotide resolution of methylation patterns.
• Analyzing complex sequencing data to identify differential methylation presents significant computational challenges.

Purpose of the Study:

• To develop an efficient and accurate method for identifying differentially methylated cytosines (dmcs).
• To address the challenges posed by complex, high-dimensional sequencing data in methylation analysis.

Main Methods:

• Development of DMCFB, a novel method employing Bayesian functional regression for dmc identification.
• Performance evaluation using simulations to compare DMCFB against existing methods.
• Application of DMCFB to analyze DNA methylation data from acute promyelocytic leukemia patients and controls.

Main Results:

• DMCFB demonstrated superior performance compared to current methods in simulations, offering better data smoothing and imputation.
• Analysis of acute promyelocytic leukemia data revealed numerous novel dmcs.
• Enhanced consistency of differential methylation was observed within CpG islands and their shores.
• Differential methylation was detected at additional binding sites of the key fused gene implicated in the cancer.

Conclusions:

• DMCFB is an efficient and effective tool for identifying differentially methylated cytosines.
• The method enhances the discovery and consistency of differential methylation patterns.
• Findings provide new insights into the epigenetic landscape of acute promyelocytic leukemia, particularly concerning key gene targets.