Splice Junction Detection

  • Splice junction detection is an important process in transcriptomics and bioinformatics used to identify the boundaries where two exons are joined together after introns are removed from precursor messenger RNA (pre-mRNA). During RNA splicing, intronic sequences are removed and neighboring exons are connected to produce mature RNA transcripts. The genomic positions where these exon–exon connections occur are called splice junctions. Detecting splice junctions helps researchers identify gene structures, characterize alternative splicing, discover previously unknown transcripts, and understand how gene expression is regulated in different tissues, developmental stages, and biological conditions.
  • In eukaryotic organisms, most protein-coding genes contain exons separated by introns. The spliceosome, a molecular complex consisting of small nuclear ribonucleoproteins (snRNPs) and associated proteins, recognizes specific RNA sequences and removes introns through two transesterification reactions. The major spliceosome primarily recognizes introns beginning with GU at the 5′ splice site and ending with AG at the 3′ splice site. These conserved sequences, together with the branch-point sequence and polypyrimidine tract, help determine accurate intron removal. After splicing, the upstream and downstream exons become connected, creating an exon–exon junction that can be detected through RNA sequencing.
  • Splice junctions provide direct evidence of RNA processing and transcript structure. A splice junction is generally represented by the genomic coordinate of the end of one exon and the beginning of the next exon, together with the strand and supporting sequencing reads. For example, if an intron lies between two exons, reads that span the exon–exon boundary indicate that the intron has been removed from the RNA molecule. These reads are commonly called junction-spanning reads or split reads because different portions of a single sequencing read map to separate genomic regions.
  • RNA sequencing (RNA-seq) is one of the most widely used methods for splice junction detection. During RNA-seq, RNA molecules are converted into complementary DNA (cDNA), fragmented, and sequenced. The resulting reads are aligned against a reference genome or transcriptome. Reads that cross exon–exon boundaries cannot usually be aligned as one continuous segment to the genomic reference because the corresponding intron is absent from the mature RNA. Splice-aware aligners therefore divide these reads into segments and map each segment to the appropriate exon. The gap between the mapped segments represents the intron that was removed during RNA processing, while the connection between the segments identifies the splice junction.
  • Splice-aware alignment is essential for accurate splice junction detection. Conventional alignment algorithms designed for genomic DNA may fail to correctly map reads spanning exon–exon boundaries. Specialized tools such as STAR, HISAT2, and TopHat2 were developed to identify spliced alignments. STAR uses a fast sequential search strategy to identify splice junctions across the genome, while HISAT2 uses a hierarchical indexing approach to support efficient alignment. TopHat2 was historically important in RNA-seq analysis but is now generally considered an older tool; STAR and HISAT2 are more commonly used in modern workflows. These aligners can identify both annotated junctions, which are already present in reference gene annotations, and novel junctions that are supported by sequencing data but are not included in existing annotations.
  • A typical splice junction detection workflow begins with RNA-seq quality assessment using FastQC and MultiQC. Low-quality bases and adapter sequences can be removed using Cutadapt or fastp. Clean reads are then aligned to a reference genome using a splice-aware aligner such as STAR or HISAT2. The resulting alignment files, commonly stored in BAM or CRAM format, contain information about read mapping positions and split alignments. Junction information can be extracted from aligner-specific output files or from alignment records using tools such as regtools, samtools, and other RNA-seq processing utilities. The detected junctions can then be compared with reference annotations from Ensembl, RefSeq, or GENCODE.
  • Splice junction detection is closely connected to alternative splicing analysis. A gene may use different splice junctions to produce multiple transcript isoforms. For example, exon skipping produces a junction connecting the exon before the skipped exon directly to the exon after it. Alternative 5′ splice-site selection and alternative 3′ splice-site selection produce junctions with different exon boundaries. Intron retention generally does not produce a conventional exon–exon junction across the retained intron, although other junctions within the same transcript may still be detected. Mutually exclusive exon events, alternative first exons, and alternative last exons can also produce different junction patterns. By comparing junction usage between samples, researchers can identify changes in transcript processing and gene regulation.
  • Splice junction support is commonly measured by counting the number of sequencing reads that span a particular exon–exon boundary. Junction read counts provide information about the relative usage of different splice sites, although the number of supporting reads is influenced by gene expression, sequencing depth, read length, mapping quality, and library preparation. A simplified representation of junction read support is:
    • Junction Support = Number of Reads Spanning the Exon–Exon Junction
  • For a gene with multiple alternative junctions, the relative usage of a junction can be estimated by comparing its supporting reads with the total supporting reads for competing junctions. A simplified expression is:
    • Junction Usage = Reads Supporting a Junction / Total Reads Supporting Competing Junctions
  • Here, reads supporting a junction represent sequencing reads that cross a specific exon–exon boundary, while total reads supporting competing junctions represent the combined evidence for alternative connections in the same splicing event. This is a simplified measure. Actual methods may account for effective junction length, read position, transcript abundance, normalization, and other factors. Junction usage should therefore be interpreted according to the method used.
  • Novel splice junction detection is particularly important because existing genome annotations do not contain every transcript expressed in an organism. RNA-seq can reveal previously unknown exon combinations, alternative splice sites, and transcripts expressed only in specific tissues or conditions. However, a novel junction should not automatically be considered a genuine biological event. False junctions may arise from sequencing errors, low-complexity sequences, repetitive regions, paralogous genes, incorrect genome assembly, or alignment artifacts. Reliable identification requires sufficient supporting reads, appropriate mapping quality, canonical splice-site signals where expected, consistent evidence across biological replicates, and careful comparison with genomic annotations.
  • Several computational tools are used for splice junction analysis. STAR and HISAT2 perform splice-aware read alignment and provide junction information. Regtools can extract and process splice junctions from aligned RNA-seq reads. StringTie uses alignment evidence to assemble and quantify transcript structures. DEXSeq can investigate differential exon usage, while rMATS, SUPPA2, MAJIQ, and LeafCutter are used to analyze alternative splicing events and changes in splice junction usage. Junction-based approaches are particularly useful for detecting complex alternative splicing patterns that may be difficult to identify using gene-level expression measurements alone.
  • Splice junction detection can also be used to investigate differential splicing between biological conditions. For example, researchers may compare junction usage between healthy and diseased tissues, different developmental stages, treated and untreated cells, or different animal breeds. Changes in junction usage can indicate alternative exon inclusion, exon skipping, or alternative splice-site selection. Statistical methods are used to determine whether observed differences are consistent across biological replicates and are unlikely to result from random sampling variation. The interpretation of differential splicing should consider gene expression changes, sequencing depth, sample composition, and possible technical batch effects.
  • Long-read sequencing technologies, including Oxford Nanopore and PacBio sequencing, provide additional opportunities for splice junction analysis. Short-read sequencing often detects individual junctions without revealing which distant exons belong to the same complete transcript. Long reads can span multiple exons and junctions within individual RNA molecules, making it easier to reconstruct full-length transcript isoforms and identify combinations of alternative splicing events. Long-read sequencing is particularly useful for genes with complex transcript structures, although sequencing accuracy, transcript-end detection, and computational processing remain important considerations.
  • Functional interpretation of splice junctions involves examining how alternative exon connections influence the resulting RNA and protein products. A splice junction may preserve the reading frame, introduce a frameshift, remove an important protein domain, or create a premature termination codon. Transcripts containing premature termination codons may be degraded through nonsense-mediated mRNA decay (NMD). Functional annotation can be performed using resources such as Ensembl, RefSeq, UniProt, InterPro, Gene Ontology (GO), and KEGG. Integrating splice junction results with protein-domain analysis, gene expression, and pathway analysis can help explain the biological consequences of alternative splicing.
  • Splice junction detection is also important in disease research. Abnormal splicing can result from mutations in splice-site sequences, changes in splicing-factor expression, or defects in spliceosome components. Such changes may contribute to cancer, neurological disorders, inherited genetic diseases, and other conditions. RNA-seq-based junction analysis can help identify abnormal exon connections, disease-associated transcript isoforms, and potential molecular biomarkers. In some cases, splice junctions can also provide information for developing therapeutic strategies, including antisense oligonucleotides designed to modify specific splicing events.
  • In animal genomics and breeding research, splice junction detection can help explain variation in gene regulation and transcript structure among breeds, tissues, developmental stages, and environmental conditions. RNA-seq studies can identify junctions associated with genes involved in growth, muscle development, milk production, reproductive performance, feed efficiency, immune function, and disease resistance. For example, alternative junction usage in muscle-related genes may contribute to differences in protein structure and muscle development, while changes in immune-related transcripts may be associated with disease responses. Such findings can support the identification of candidate regulatory mechanisms and functional genetic variants, although their relevance to breeding decisions requires further validation.
  • Experimental validation is essential when novel or biologically important splice junctions are identified. Reverse transcription PCR (RT-PCR) can use primers positioned in different exons to amplify a specific exon–exon junction. The resulting products can be separated by gel electrophoresis and sequenced to confirm the expected exon connection. Quantitative PCR (qPCR) can measure relative expression of specific splice isoforms when junction-specific primers are available. Sanger sequencing can confirm the precise junction sequence, while long-read sequencing can provide additional evidence about the complete transcript structure.
  • Overall, splice junction detection is a fundamental bioinformatics approach for understanding RNA processing, transcript structure, and alternative splicing. RNA-seq combined with splice-aware alignment, junction extraction, transcript reconstruction, statistical analysis, and functional annotation enables researchers to identify known and novel exon–exon connections. Integrating short-read and long-read sequencing with experimental validation improves the reliability of transcript identification. Splice junction detection is therefore essential in transcriptomics, functional genomics, disease research, and animal genomics, particularly when investigating how alternative RNA processing contributes to gene expression and biological variation.
Author:

Leave a Reply

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