PennSeq: accurate isoform-specific gene expression quantification in RNA-Seq by modeling non-uniform read...

Area of Science:

• Genomics
• Bioinformatics
• Computational Biology

Background:

• Accurate isoform-specific gene expression estimation is crucial for understanding biological mechanisms and disease genetics.
• RNA sequencing (RNA-Seq) data analysis is complicated by various biases that can affect expression estimation.
• Existing methods may struggle with non-uniform read distributions inherent in RNA-Seq data.

Purpose of the Study:

• To introduce PennSeq, a statistical method for estimating isoform-specific gene expression.
• To address challenges posed by non-uniform read distributions in RNA-Seq data.
• To provide a robust and accurate tool for gene expression analysis.

Main Methods:

• Developed PennSeq, a non-parametric statistical method.
• PennSeq models isoform-specific non-uniform read distributions empirically from aligned data.
• The method accounts for various biases like hexamer priming, local sequence, positional effects, and RNA degradation.

Main Results:

• PennSeq demonstrated superior performance compared to existing methods in simulations and real-world data.
• The method showed particular effectiveness for isoforms with severe non-uniformity.
• Evaluated using simulated data and two real Illumina RNA-Seq datasets, including one with qPCR validation.

Conclusions:

• PennSeq offers a powerful, data-driven approach to isoform expression estimation.
• The method effectively corrects for biases leading to non-uniform read distributions.
• PennSeq enhances the accuracy of gene expression analysis, aiding biological and disease gene discovery.