flexiMAP: a regression-based method for discovering differential alternative polyadenylation events in standard...

Area of Science:

• Genomics
• Bioinformatics

Background:

• Alternative polyadenylation (APA) is a key mechanism regulating gene expression.
• Accurate detection of differential APA events is crucial for understanding biological processes.
• Existing methods have limitations, particularly with low fold-change events and confounding variables.

Purpose of the Study:

• To introduce flexiMAP, a novel beta-regression-based method for differential APA event discovery.
• To evaluate flexiMAP's performance against existing methods using simulated and real RNA-seq data.
• To provide a flexible tool that can incorporate known covariates in APA analysis.

Main Methods:

• Development of flexiMAP, a new method implemented in R.
• Utilizing beta-regression for modeling APA events.
• Comparison with existing methods on simulated and real RNA-seq datasets.
• Incorporation of multiple known covariates to adjust for confounding factors.

Main Results:

• flexiMAP demonstrates a favorable balance of specificity and sensitivity.
• The method outperforms existing approaches, particularly for detecting small fold-change APA events.
• Simulated data analysis revealed previously unrecognized limitations of current methods.
• flexiMAP effectively models the influence of known covariates on APA analysis.

Conclusions:

• flexiMAP is a robust and sensitive tool for differential APA event detection in RNA-seq data.
• The R package provides a valuable resource for researchers studying gene expression regulation.
• flexiMAP's ability to handle covariates enhances the reliability of APA analysis.