RUM

RUM maps short RNA-Seq reads to reference genomes, accounting for RNA splicing to produce accurate alignments for transcriptome analysis.

Key Features:

• BLAT-based alignment: Implements a pipeline based on BLAT to align RNA-Seq reads to reference genomes.
• Splice-aware alignment: Handles spliced alignments and detection of splice junctions to accommodate RNA splicing.
• Integrated RNA-Seq simulator: Includes a simulator that models alternative splicing, insertions, deletions, substitutions, sequencing errors, and intron signals.
• Benchmarking capability: Enables controlled benchmarking of alignment methods with simulated reads to assess per-base and splice-junction accuracy.
• Comparative assessment: Compares performance against established and novel aligners to evaluate accuracy and speed trade-offs.
• Transcript discovery: Detects novel transcript features such as new exons and alternative splicing events.
• Public validation dataset: Development and validation used RNA-Seq reads from mouse retina available at GEO accession GSE26248.

Scientific Applications:

• Transcript discovery: Identification of novel exons and alternative splicing events in RNA-Seq data.
• Alignment benchmarking: Evaluation and comparison of RNA-Seq alignment algorithms using simulated and empirical reads.
• Accuracy assessment: Measurement of per-base and splice-junction alignment accuracy for aligner evaluation.
• Experimental validation support: Generation of candidate transcript features for validation by RT-PCR and Sanger sequencing (as applied to mouse retina data).

Methodology:

RUM uses a BLAT-based alignment pipeline and an integrated RNA-Seq simulator that models alternative splicing, insertions, deletions, substitutions, sequencing errors, and intron signals, and it performs comparative benchmarking of aligners by assessing base-level and splice-junction accuracy.