- Copy-number variants (CNVs) and structural variants (SVs) are important forms of genetic variation involving changes in the organization, length, or copy number of genomic DNA. Unlike single-nucleotide variants, which affect individual nucleotide positions, these variants involve larger DNA segments and may influence multiple genes, regulatory regions, or entire chromosomal regions. They contribute to genetic diversity, genome evolution, inherited disorders, cancer development, population variation, and differences in important biological traits. Their detection and interpretation are major components of modern genomic analysis.
- Structural variants are generally defined as genomic alterations involving DNA segments that are larger or more complex than typical small insertions and deletions. They include deletions, duplications, insertions, inversions, translocations, and complex rearrangements. The size threshold used to distinguish structural variants from small variants varies among research studies and bioinformatics tools. A threshold of approximately 50 base pairs is commonly used in many sequencing-based workflows, but this is a convention rather than a universal biological boundary. Some tools identify smaller events, while others focus on much larger genomic changes.
- Copy-number variants are a major category of structural variation in which the number of copies of a particular genomic region differs among individuals or cells. A region may be deleted, resulting in fewer copies than expected, or duplicated, resulting in additional copies. In a typical diploid human genome, most autosomal regions have two copies, one inherited from each parent. A deletion may reduce the copy number to one, while a duplication may increase it to three or more. However, copy-number interpretation depends on chromosome ploidy, sex chromosomes, sample type, and the biological characteristics of the organism.
- Structural variants and copy-number variants are closely related but are not interchangeable terms. Structural variation describes a broad range of changes in genomic organization, including alterations that do not necessarily change the total copy number of a region. For example, an inversion changes the orientation of a DNA segment without necessarily changing its copy number. A balanced translocation moves DNA between chromosomes without necessarily producing a net gain or loss of genetic material. CNVs, by contrast, specifically involve gains or losses of genomic copies. Many deletions and duplications are therefore both structural variants and copy-number variants.
- Deletions occur when a segment of DNA is lost from the genome. Their size may range from a few bases to millions of nucleotides. A deletion may remove part of a gene, an entire gene, several neighbouring genes, or a large chromosomal region. The biological consequences depend on the affected sequence, the number of copies remaining, the organism’s ploidy, and whether the deleted region contains essential genes or regulatory elements. Deletions may disrupt protein-coding sequences, change gene dosage, or remove regulatory elements that control nearby genes.
- Duplications occur when a DNA segment is copied and an additional copy is introduced into the genome. They can involve part of a gene, an entire gene, or larger genomic regions. Gene duplication is an important mechanism in genome evolution because duplicated genes may retain the original function, acquire new functions, or undergo functional specialization. However, additional gene copies can also disturb normal gene dosage and contribute to disease or abnormal development. Some duplications are tandem, with the extra copy located adjacent to the original sequence, while others may occur at different genomic positions.
- Insertions are structural changes in which additional DNA sequences are introduced into a genomic location. These sequences may originate from another genomic region, mobile genetic elements, duplicated DNA, or other molecular processes. Transposable elements, including LINEs, SINEs, and other mobile elements, can create insertional variation by moving or copying themselves into new genomic locations. Such insertions may disrupt genes, alter regulatory activity, or influence genome structure. Some insertions are difficult to detect using short-read sequencing because the inserted sequence may not be present in the reference genome.
- Inversions occur when a segment of DNA is reversed in orientation within a chromosome. A segment that originally follows one sequence direction is rearranged so that it appears in the opposite orientation. Inversions can be classified as paracentric when they do not include the centromere, or pericentric when they include the centromere. Although an inversion may not change the total amount of DNA, it can affect gene function if a breakpoint disrupts a gene or regulatory region. Inversions may also suppress recombination in heterozygous regions and influence the inheritance of linked genetic variants.
- Translocations involve the movement or exchange of DNA segments between different chromosomal locations. Reciprocal translocations involve the exchange of segments between two chromosomes, while non-reciprocal translocations involve movement in one direction without an equivalent exchange. Some translocations are balanced, meaning that there is no obvious net gain or loss of genetic material. Others are unbalanced and produce duplications or deletions. Translocations may disrupt genes, alter regulatory relationships, or create fusion genes. Certain chromosomal translocations are well established in cancer biology because they can generate abnormal proteins or change gene expression.
- Complex structural variants involve multiple genomic rearrangements that cannot be described adequately as a single simple deletion, duplication, inversion, or translocation. They may contain combinations of gains, losses, inversions, insertions, and rearranged DNA segments. Complex events can arise through chromosome breakage and inaccurate repair, replication-based mechanisms, or other processes involving multiple DNA rearrangements. Some complex structural variants are associated with developmental disorders, cancer, genomic instability, and evolutionary changes.
- Copy-number variation can be described at several levels. A copy-number gain indicates that a genomic region has more copies than expected, while a copy-number loss indicates fewer copies. A region may show a simple gain or loss, a high-level amplification, or a more complex pattern involving multiple copy-number states. Amplification refers to an increase in the number of copies of a DNA segment, sometimes to a very high level. It is particularly important in cancer genomics, where amplification of oncogenes can contribute to tumour growth and progression.
- The biological effects of CNVs and SVs depend on their genomic location, size, gene content, inheritance pattern, and surrounding genomic context. A deletion involving a dosage-sensitive gene may cause disease because one functional copy is insufficient to maintain normal biological activity. A duplication may increase gene expression and disrupt normal cellular processes. Some variants may affect multiple genes simultaneously, producing a broad range of phenotypic effects. Other variants may have little detectable effect or may contribute to normal variation within a population.
- Structural variants can influence gene function through several mechanisms. A breakpoint may interrupt a protein-coding gene, alter an exon, or disrupt a regulatory sequence. A rearrangement may separate a gene from its normal enhancer or bring it into contact with a different regulatory region. Such changes can modify gene expression even when the coding sequence remains intact. Structural variation can also influence chromatin organization, gene dosage, recombination, and the activity of regulatory elements over large genomic distances.
- The formation of CNVs and SVs involves several molecular mechanisms. Non-allelic homologous recombination occurs when similar DNA sequences at different genomic locations recombine incorrectly, producing deletions, duplications, or other rearrangements. Non-homologous end joining repairs DNA double-strand breaks by joining broken ends, sometimes introducing small insertions or deletions at the junction. Microhomology-mediated repair uses short matching sequences near DNA breakpoints and can produce complex rearrangements. Replication-based mechanisms, including fork stalling and template switching, can also generate duplications and complex copy-number changes.
- Mobile genetic elements are another source of structural variation. Transposable elements can insert into new genomic locations and may contribute to deletions, duplications, and rearrangements. Recombination between repeated elements can produce changes in chromosome structure. These mechanisms are particularly relevant in genomes containing large numbers of repetitive sequences. The contribution of each mechanism varies among species, genomic regions, and biological conditions.
- CNVs and SVs can be inherited through the germline or acquired in somatic cells. Germline variants occur in reproductive cells or their precursors and may be transmitted to offspring. They contribute to inherited genomic diversity and may be associated with congenital disorders, developmental abnormalities, or inherited susceptibility to disease. Somatic structural variants arise in body cells and may accumulate during development, ageing, or tumour formation. They are particularly important in cancer because chromosome rearrangements, gene amplifications, and deletions can alter pathways controlling cell growth, survival, and genome stability.
- The detection of CNVs and SVs requires methods capable of identifying changes in DNA sequence, read alignment, chromosome structure, or copy number. Whole-genome sequencing provides broad coverage and can detect many classes of structural variation. Short-read sequencing is widely used for genome-wide analysis, while long-read sequencing can help resolve complex rearrangements and repetitive regions. Whole-exome sequencing can identify some copy-number changes affecting captured coding regions, but it provides limited information about noncoding structural variation and breakpoint locations. Targeted sequencing can examine selected regions but may not detect rearrangements outside the targeted areas.
- Chromosomal microarray analysis is a widely used method for detecting copy-number changes across the genome. Comparative genomic hybridization arrays and single-nucleotide polymorphism arrays can identify deletions and duplications by measuring relative signal intensity across genomic regions. SNP arrays can also provide information about genotype patterns and regions of homozygosity. However, arrays generally cannot identify all balanced rearrangements, and their ability to detect small or complex variants depends on probe density and platform design.
- Read-depth analysis is one of the principal computational approaches for detecting CNVs from sequencing data. It examines the number of sequencing reads aligned to different genomic regions. A region with fewer reads than expected may indicate a deletion, while a region with increased read depth may indicate a duplication or amplification. Read-depth methods require careful normalization because sequencing coverage can vary due to GC content, mappability, library preparation, and other technical factors. Tools such as CNVkit, Control-FREEC, and GATK gCNV can be used for copy-number analysis in suitable datasets.
- Paired-end mapping provides another method for detecting structural variants. In paired-end sequencing, both ends of a DNA fragment are sequenced, and the expected distance and orientation between the reads are known. If the mapped reads show an unexpected orientation or separation, this may indicate an insertion, deletion, inversion, or translocation. Discordant read pairs provide indirect evidence of structural variation, although repetitive regions and mapping ambiguity can complicate interpretation.
- Split-read analysis identifies sequencing reads that align partly to one genomic region and partly to another. Such reads can provide evidence for structural-variant breakpoints because a single read may span the junction between two rearranged DNA segments. Split-read methods can help identify precise breakpoint positions, particularly when the read length is sufficient and the breakpoint region is uniquely mappable. Their effectiveness depends on sequencing depth, read length, error rate, and the complexity of the rearrangement.
- Assembly-based approaches identify structural variation by assembling sequencing reads into longer genomic sequences and comparing the resulting assemblies with a reference genome or with one another. These approaches can be particularly useful for detecting insertions and complex rearrangements that are difficult to characterize through read alignment alone. Long-read sequencing technologies, including PacBio HiFi and Oxford Nanopore sequencing, can provide reads spanning large genomic regions and may improve the resolution of repetitive and structurally complex loci. Haplotype-resolved assemblies can further distinguish variants occurring on different chromosome copies.
- Several bioinformatics tools are used for structural-variant discovery. Manta detects structural variants from paired-end and split-read evidence in short-read sequencing data. DELLY identifies structural variants using paired-end and split-read information. GRIDSS uses genome-wide breakend assembly to identify structural rearrangements. LUMPY combines different types of alignment evidence to detect structural variation. For long-read data, Sniffles and cuteSV are commonly used to identify insertions, deletions, and other structural variants. Tool performance varies with sequencing technology, coverage, genome complexity, and the type of variant being investigated.
- For CNV analysis, CNVkit is commonly used with whole-exome and whole-genome sequencing data. GATK gCNV provides methods for detecting germline copy-number variation using sequencing read counts. Control-FREEC estimates copy number and can account for factors such as GC-content variation and tumour purity in appropriate analyses. Other tools, including CNVnator, ExomeDepth, and XHMM, are used in particular sequencing contexts. The choice of tool depends on whether the data are from whole-genome sequencing, exome sequencing, targeted sequencing, or another platform.
- Structural variants are commonly represented using specialized fields and records in the Variant Call Format. Simple deletions and insertions can often be represented directly through the REF and ALT alleles. Larger or more complex events may require symbolic alleles, breakend notation, and additional metadata. For example, a symbolic deletion may be represented as:
##fileformat=VCFv4.3
##reference=GRCh38
#CHROM POS ID REF ALT QUAL FILTER INFO
chr1 123456 DEL001 N <DEL> 60 PASS SVTYPE=DEL;END=124456;SVLEN=-1000
- This illustrative record describes a deletion on chromosome 1 with a reported length of 1,000 bases. The symbolic allele <DEL> indicates a deletion, while END identifies the end coordinate and SVLEN describes the length of the event. A complete VCF must include appropriate metadata declarations for the fields used. Exact coordinate and length conventions should be checked against the VCF specification and the software that generated the record.
- The VCF specification also provides breakend notation for representing rearrangements in which genomic segments are connected in different orientations or at different locations. Breakend records can describe complex events and reciprocal relationships between genomic positions. Because structural variants may involve multiple breakpoints, a single biological event may be represented by more than one VCF record. Researchers should interpret related records together and examine the caller’s documentation to understand how events are encoded.
- Quality assessment is essential for reliable CNV and SV detection. Important evidence includes read depth, mapping quality, split-read support, discordant read pairs, breakpoint consistency, variant size, and the number of independent supporting reads. For CNVs, normalized coverage, signal intensity, B-allele frequency, and segmentation quality may also be considered. In cancer samples, tumour purity and subclonal composition can influence estimated copy number. Variants supported by weak or inconsistent evidence may represent technical artifacts, so confidence assessment and validation are important.
- CNV and SV filtering strategies depend on the data type and analysis objective. Researchers may apply minimum support thresholds, quality scores, read-depth requirements, size limits, and genomic-region filters. Variants in low-complexity regions, segmental duplications, and poorly mappable regions often require additional scrutiny. Some analyses prioritize variants that affect known disease-associated genes, while others retain a broad set of variants for population or evolutionary studies. Filtering should be designed carefully because overly strict criteria can remove genuine variants, while weak criteria can retain false positives.
- After detection, CNVs and SVs are annotated to determine which genes, transcripts, regulatory elements, and genomic regions are affected. Annotation can identify whether a variant overlaps a coding sequence, disrupts a gene, changes gene dosage, or affects a known regulatory region. Tools such as Ensembl Variant Effect Predictor, AnnotSV, and other structural-variant annotation systems can support this process. Interpretation may also involve databases containing known structural variants, gene-disease associations, population frequencies, and clinical classifications.
- In clinical genetics, CNVs and SVs are important causes of inherited disorders and developmental conditions. Deletions may remove one or more genes, while duplications may increase gene dosage. Some structural rearrangements can disrupt genes or produce abnormal fusion genes. Clinical interpretation considers the size and location of the variant, the genes and regulatory regions affected, inheritance pattern, population frequency, phenotype, and available functional evidence. Resources such as ClinGen and ClinVar can support interpretation, but the evidence behind each record should be reviewed rather than relying solely on a database classification.
- In cancer genomics, structural variants and CNVs can contribute to tumour development, progression, and treatment resistance. Gene amplification may increase the expression of oncogenes, while deletions may remove tumour-suppressor genes. Chromosomal translocations can generate fusion genes, and complex rearrangements can alter several cancer-related pathways. Copy-number profiles are also used to characterize tumour genomes and investigate genomic instability. Because cancer samples may contain mixtures of tumour and normal cells, copy-number analysis must consider tumour purity, ploidy, and intratumour heterogeneity.
- In population genetics and evolutionary biology, CNVs and SVs contribute to genetic diversity and genome evolution. They can influence gene content, gene dosage, adaptation, and differences among populations or species. Some gene duplications provide raw material for evolutionary innovation, while deletions may remove genes that are no longer advantageous or are functionally redundant. Structural variants can also affect recombination and the inheritance of linked alleles. Population-level studies use these variants to investigate demographic history, selection, ancestry, and genome organization.
- In agriculture and animal breeding, CNVs and SVs can influence economically important traits such as growth, fertility, milk production, meat quality, disease resistance, feed efficiency, and environmental adaptation. In livestock, copy-number differences may affect genes involved in production and physiological functions. In plants, structural variation can contribute to differences in flowering time, stress tolerance, disease resistance, yield, and product quality. These variants can be incorporated into genetic studies and breeding programmes when their effects are sufficiently established and validated in relevant populations.
- Microbial genomes also contain structural variation, including deletions, duplications, insertions, inversions, plasmid-related changes, and mobile-element activity. Such variation can contribute to antimicrobial resistance, virulence, adaptation, and differences among microbial strains. Bacterial pangenome analysis is particularly useful because it distinguishes genes shared across a species from accessory genes present in only some strains. Structural changes may alter gene content or regulatory relationships, making them important for microbial evolution and epidemiology.
- Despite advances in sequencing and computational analysis, CNV and SV detection remains challenging. Short-read sequencing may not resolve repetitive regions or complex breakpoints, while read-depth methods can be affected by coverage variation and GC bias. Long-read sequencing improves the characterization of many complex events but may require higher sequencing costs or specialized analysis. Different tools may produce different calls because they use different evidence types and algorithms. Combining read-depth, paired-end, split-read, and assembly-based evidence can improve confidence, but the integration process must be carefully designed.
- Validation is important for high-confidence structural-variant findings, especially when variants have potential clinical or breeding significance. Depending on the variant and research objective, validation may use PCR, quantitative PCR, digital PCR, Sanger sequencing of breakpoints, chromosomal microarray analysis, fluorescence in situ hybridization, or another appropriate method. No single validation method is suitable for every structural variant. The chosen method should be capable of confirming the specific type, size, and genomic location of the event.
- Reproducible CNV and SV analysis requires documentation of the reference genome assembly, sequencing technology, read-processing procedures, alignment software, variant caller, software versions, parameters, and filtering criteria. Researchers should preserve raw sequencing data, alignment files, variant-call files, coverage information, and analysis scripts. Variant coordinates must be linked to the correct reference assembly, and structural-variant representations should be checked before comparison with external databases. When converting between reference assemblies, some variants may fail to map or may map ambiguously, particularly in regions with complex structural differences.
- Copy-number variants and structural variants are essential components of genomic diversity. They can alter gene dosage, disrupt genes, change regulatory activity, reorganize chromosomes, and influence biological traits and disease susceptibility. Their analysis requires specialized sequencing and computational methods, careful quality assessment, accurate reference representation, and appropriate functional interpretation. As long-read sequencing, pangenome resources, and graph-based genome representations continue to develop, the detection and understanding of structural variation will become increasingly comprehensive and important across medical, evolutionary, agricultural, and microbial genomics.