iREAD

iREAD detects intron retention (IR) events from RNA-seq data to enable genome-wide identification of retained introns implicated in gene expression regulation and disease.

Key Features:

• Input files: Accepts a BAM file of aligned RNA-seq reads and a text file of independent intron coordinates.
• Independent introns definition: Uses introns defined as not overlapping any exons of splice isoforms.
• Read counting: Counts reads that overlap intronic regions in the BAM file.
• Read feature analysis: Analyzes intronic read features including read depth and within-intron read distribution patterns.
• Filtering criteria: Applies thresholds on number of reads/fragments in introns, FPKM, junction reads, and intronic read distribution patterns to select reliable IR events.
• Output: Produces a tab-delimited text file listing retained introns for downstream analyses.

Scientific Applications:

• Genome-wide intron retention profiling: Enables identification of retained introns across the transcriptome from RNA-seq data.
• Differential intron expression studies: Provides retained-intron lists suitable for comparing intron retention between conditions.
• Functional enrichment and regulatory analysis: Supplies input for functional enrichment to infer regulatory mechanisms and disease associations linked to IR.
• Analysis of poly(A)-enriched transcriptomes: Applicable to RNA-seq libraries enriched for polyadenylated RNA.
• Comparative evaluation with IRFinder: Reported higher Area Under the Curve (AUC) than IRFinder across datasets and, together with IRFinder, can complement detection due to low false-positive rates and variable false-negative rates.

Methodology:

Counts reads overlapping intronic regions from a BAM file, analyzes read depth and within-intron read distribution including junction reads, computes FPKM and applies thresholds on read/fragment counts, FPKM, junction reads, and intronic read distribution patterns to output a tab-delimited list of retained introns.