Overview
STAR (Spliced Transcripts Alignment to a Reference) is a standalone aligner specifically designed for RNA-seq data. It employs a seed-based approach, enabling ultrafast and accurate alignment of RNA-seq reads to a reference genome. STAR excels at detecting splice junctions and handling large mammalian genomes with substantial memory requirements (16-32GB RAM). Its architecture involves two main stages: seed searching and spliced alignment. It utilizes maximal mappable prefix (MMP) search to efficiently identify potential alignment locations, followed by precise spliced alignment using a dynamic programming approach. Use cases include gene expression quantification, isoform discovery, and variant calling in RNA-seq experiments.