A powerful and flexible statistical framework for testing hypotheses of allele-specific gene expression from RNA-seq...

Area of Science:

• Genomics
• Statistical Genetics
• Molecular Biology

Background:

• Gene expression variation contributes to phenotypic diversity.
• RNA sequencing (RNA-seq) data offers insights into regulatory variation.
• Novel statistical methods are needed to analyze complex RNA-seq data.

Purpose of the Study:

• To develop a statistical framework for analyzing allele-specific expression (ASE) using RNA-seq data.
• To enable both global and locus-specific inferences of ASE.
• To provide a high-resolution tool for profiling genome-wide ASE.

Main Methods:

• Developed a hierarchical Bayesian model to combine information across genetic loci.
• Applied the model to RNA-seq data from a diploid hybrid yeast and a human genome.
• Quantified ASE levels with controlled false-discovery rates.

Main Results:

• The methodology accurately quantifies ASE and achieves high reproducibility across sequencing platforms.
• Identified loci with significant and biologically relevant patterns of ASE.
• Detected allele-specific alternative splicing and transcription termination events.

Conclusions:

• The developed Bayesian model is a rigorous and quantitative tool for ASE profiling.
• This approach enhances the understanding of regulatory variation and its contribution to phenotypic diversity.
• The methodology is applicable to diverse organisms, including yeast and humans.