Detecting m6A methylation regions from Methylated RNA Immunoprecipitation Sequencing

Motivation: The post-transcriptional epigenetic modification on mRNA is an emerging field to study the gene regulatory mechanism and their association with diseases. Recently developed high-throughput sequencing technology named Methylated RNA Immunoprecipitation Sequencing (MeRIP-seq) enables one to profile mRNA epigenetic modification transcriptome wide. A few computational methods are available to identify transcriptome-wide mRNA modification, but they are either limited by over-simplified model ignoring the biological variance across replicates or suffer from low accuracy and efficiency. Results: In this work, we develop a novel statistical method, based on an empirical Bayesian hierarchical model, to identify mRNA epigenetic modification regions from MeRIP-seq data. Our method accounts for various sources of variations in the data through rigorous modeling and applies shrinkage estimation by borrowing information from transcriptome-wide data to stabilize the parameter estimation. Simulation and real data analyses demonstrate that our method is more accurate, robust and efficient than the existing peak calling methods.

Automated summary

In this work, a novel statistical method is developed to identify mRNA epigenetic modification regions from MeRIP-seq data, based on an empirical Bayesian hierarchical model. This method accounts for various sources of variations in the data through rigorous modeling and applies shrinkage estimation by borrowing information from transcriptome-wide data to stabilize the parameter estimation, showing higher accuracy, robustness, and efficiency compared to existing peak calling methods in simulation and real data analyses.