An Island-Based Approach for Differential Expression Analysis.

Area of Science:

• Genomics
• Bioinformatics
• Transcriptomics

Background:

• High-throughput mRNA sequencing (RNA-Seq) is crucial for transcriptome profiling, enabling gene expression quantification and novel transcript discovery.
• Current RNA-Seq methods rely on annotated genomic features, limiting the detection of expression changes in unannotated regions.
• This limitation hinders comprehensive analysis of differential gene expression.

Purpose of the Study:

• To develop a novel segmentation approach for analyzing differential expression in RNA-Seq and targeted sequencing data.
• To overcome the limitations of existing methods that require prior knowledge of genomic annotations.
• To enable the detection of differential expression in both annotated and unannotated genomic regions.

Main Methods:

• Developed the Island-Based (IB) segmentation approach for RNA-Seq data analysis.
• IB determines expression islands based on windowed read counts for cross-experimental comparison.
• Combined island significance (p-values) using Fisher's method to detect differentially expressed genes.

Main Results:

• The IB algorithm was tested against CuffDiff, DESeq, and edgeR using MAQC RNA-Seq datasets.
• IB demonstrated superior performance compared to existing methods in both datasets.
• Performance improvement was quantified by an increased area under the receiver operating characteristic curve (auROC).

Conclusions:

• The Island-Based (IB) method offers a robust approach for differential gene expression analysis in RNA-Seq.
• IB effectively analyzes expression data without relying on specific isoform knowledge.
• This novel method enhances the accuracy and scope of transcriptome profiling studies.