STAR: A Splice-Aware Aligner for RNA-seq Data

  • STAR (Spliced Transcripts Alignment to a Reference) is a widely used bioinformatics tool for aligning RNA sequencing (RNA-seq) reads to a reference genome. It is specifically designed to handle reads that span exon-exon junctions, making it particularly useful for studying gene expression, RNA splicing, and transcript diversity. STAR is known for its high alignment speed and ability to process large volumes of RNA-seq data efficiently.
  • RNA sequencing is used to study gene expression by analysing RNA molecules produced by cells and tissues. Before sequencing, RNA is typically converted into complementary DNA (cDNA), which is then sequenced to generate short reads. In eukaryotic organisms, genes contain exons and introns. During RNA processing, introns are removed through a process called splicing, and the remaining exons are joined to form mature messenger RNA (mRNA). Consequently, RNA-seq reads may contain sequences from two different exons that are separated by an intron in the reference genome. A conventional DNA read mapper may struggle to align these reads correctly because the corresponding genomic regions are not continuous.
  • STAR addresses this problem through splice-aware alignment. It can identify exon-exon junctions and align reads across introns by dividing them into smaller segments and mapping each segment to its corresponding genomic location. This process is known as split-read alignment. It allows STAR to map RNA-seq reads accurately, including those that originate from different exons of the same gene. This capability is particularly important for studying alternative splicing, in which different combinations of exons are joined to produce multiple RNA transcripts from a single gene.
  • Before mapping, STAR builds an index of the reference genome. This index is a data structure that enables the software to identify matching genomic regions quickly and efficiently. STAR uses a suffix array-based indexing approach to support rapid sequence searching. The reference genome index can be reused for multiple RNA-seq samples, provided the same compatible reference genome and indexing configuration are used. Known splice-junction information from gene annotations, such as GTF files, can also be incorporated during index generation to support more accurate alignment around annotated exon-exon junctions.
  • STAR supports both one-pass and two-pass alignment strategies. In one-pass alignment, sequencing reads are mapped to the reference genome while splice junctions are identified during the alignment process. In two-pass alignment, STAR first performs an initial mapping to identify splice junctions and then uses the detected junctions to guide a second alignment step. This approach can improve the mapping of reads spanning previously unannotated splice junctions, although two-pass alignment is optional.
  • After alignment, STAR generates output files containing information about the mapping results. The main output is usually a SAM (Sequence Alignment/Map) file, which records the genomic positions and alignment characteristics of mapped reads. These files can be converted into BAM (Binary Alignment/Map) format, which stores alignment information in a compressed binary format and is commonly used in downstream genomic analyses. STAR can also generate coordinate-sorted BAM files directly. Additional output includes alignment statistics and splice-junction information, which can help researchers evaluate mapping performance and investigate RNA splicing patterns.
  • STAR is widely used in transcriptomics research, particularly for gene expression analysis and alternative splicing studies. After mapping, tools such as featureCounts can be used to count reads assigned to individual genes, while DESeq2 can be used to identify differentially expressed genes between experimental conditions. STAR can also identify chimeric RNA alignments, which may provide evidence of gene fusion events that are particularly relevant in cancer research.
  • Unlike Bowtie 2, which is commonly used for DNA sequencing read mapping, STAR is specifically designed for RNA-seq data and supports splice-aware alignment. Although both tools build reference genome indexes to improve mapping efficiency, their primary applications differ. Bowtie 2 is suitable for mapping reads to continuous genomic regions, whereas STAR can align reads that span exon-exon junctions. However, STAR may require substantial computer memory, particularly when indexing large reference genomes, and additional tools are needed for downstream gene expression analysis.
  • In a typical RNA-seq workflow, raw sequencing reads are first checked for quality using FastQC, and reports from multiple samples can be summarised using MultiQC. Adapter sequences and low-quality bases are then removed using tools such as fastp. The cleaned reads are subsequently mapped to a reference genome using STAR. The resulting alignment files can be used for gene expression quantification, differential expression analysis, and alternative splicing studies.
  • In summary, STAR is an important bioinformatics tool for RNA-seq analysis because of its speed and ability to perform splice-aware alignment. By identifying exon-exon junctions and mapping reads across introns, it enables researchers to investigate gene expression, alternative splicing, and transcript diversity. It is therefore an essential component of many modern RNA-seq analysis workflows.
Author:

Leave a Reply

Your email address will not be published. Required fields are marked *