Copy number variations (CNVs) are a type of structural genomic variation characterized by the deletion or duplication of DNA segments ranging from about 1 kilobase (kb) to several megabases (Mb) in size. Unlike single nucleotide variants (SNVs), CNVs alter the dosage of genes and regulatory elements, directly affecting gene expression and cellular function. They account for a substantial portion of human genetic diversity and are increasingly recognized as significant contributors to both rare monogenic disorders and common complex diseases. With the advent of high-resolution genomic technologies, the landscape of CNV research has expanded dramatically, revealing their pervasive influence across the genome and their critical role in human health and disease.

Understanding Copy Number Variations

CNVs arise from non-allelic homologous recombination (NAHR), non-homologous end joining (NHEJ), and replication-based mechanisms such as fork stalling and template switching (FoSTeS) or microhomology-mediated break-induced replication (MMBIR). These events occur during meiosis or mitosis and can produce gains or losses of genomic material. Duplications can be tandem (adjacent copies) or interspersed, while deletions remove one or more copies of a segment. CNVs may encompass entire genes, partial gene sequences, or intergenic regions containing enhancers, silencers, or non-coding RNAs.

The frequency of CNVs varies across the genome; some regions—known as CNV hotspots—are prone to recurrent rearrangements due to segmental duplications (low-copy repeats). These hotspots are often associated with genomic disorders. CNVs can be inherited in a Mendelian pattern or arise de novo, especially in neurodevelopmental disorders. Population studies estimate that each individual carries hundreds of CNVs, most of which are benign. However, rare, large, or gene-disrupting CNVs can have profound phenotypic consequences. The severity of effect depends on the size, gene content, dosage sensitivity of the involved genes, and the genetic background of the individual.

Impact on Genetic Disorders

Many well-characterized genetic disorders are directly caused by specific CNVs, often referred to as genomic disorders. These conditions typically arise from deletions or duplications that alter the copy number of one or more dosage-sensitive genes. Below are several representative examples.

22q11.2 Deletion Syndrome (DiGeorge Syndrome)

One of the most common microdeletion syndromes, occurring in approximately 1 in 4,000 live births. A 1.5–3 Mb deletion on chromosome 22q11.2 removes about 30–50 genes, including TBX1, a key regulator of cardiac and parathyroid development. Clinical features include conotruncal heart defects (tetralogy of Fallot, interrupted aortic arch), thymic aplasia leading to immune deficiency, hypocalcemia due to parathyroid hypoplasia, palatal abnormalities, and characteristic facial features. Cognitive impairment, schizophrenia in later life, and autoimmune disorders are also associated. Diagnosis is confirmed by chromosomal microarray (CMA) or fluorescence in situ hybridization (FISH).

Charcot-Marie-Tooth Disease Type 1A

This hereditary peripheral neuropathy results from a 1.5 Mb duplication on chromosome 17p12 that includes the PMP22 gene. The extra copy leads to overexpression of peripheral myelin protein 22, disrupting myelin sheath formation in peripheral nerves. Symptoms include distal muscle weakness and atrophy, foot deformities (pes cavus), and sensory loss, typically beginning in the first or second decade of life. Charcot-Marie-Tooth disease type 1A accounts for about 70% of all CMT cases. Deletion of the same region causes hereditary neuropathy with liability to pressure palsies (HNPP), illustrating the dosage-sensitive nature of the locus.

Williams-Beuren Syndrome

Caused by a heterozygous deletion of ~1.5–1.8 Mb on chromosome 7q11.23, encompassing about 26–28 genes including ELN (elastin). The loss of one ELN allele results in vascular defects such as supravalvular aortic stenosis and peripheral pulmonary stenosis. Other features include characteristic "elfin" facies (wide mouth, full lips, periorbital fullness), intellectual disability with a distinctive cognitive profile (strong verbal short-term memory but poor visuospatial skills), gregarious personality, cardiovascular disease, hypercalcemia in infancy, and growth retardation. The deletion arises from NAHR between flanking low-copy repeats.

Smith-Magenis Syndrome

This microdeletion syndrome involves a 3.5 Mb deletion on chromosome 17p11.2, including the RAI1 gene. Clinical features include distinct facial features (brachycephaly, midface hypoplasia, downturned mouth), intellectual disability, expressive language delay, sleep disturbances due to inverted melatonin rhythm, self-injurious behaviors (head banging, skin picking), and obesity. RAI1 is a dosage-sensitive transcription factor; haploinsufficiency is responsible for most phenotypic features.

Neurofibromatosis Type 1

While most often caused by point mutations in NF1, a minority of cases (about 5–10%) involve a 1.2–1.5 Mb microdeletion on chromosome 17q11.2 that encompasses NF1 and adjacent genes. These patients tend to have a more severe phenotype with earlier onset, more neurofibromas, cognitive deficits, facial dysmorphism, and increased risk for malignant peripheral nerve sheath tumors. The deletion highlights the additive effect of losing multiple contiguous genes.

Additional Syndromes

  • Prader-Willi Syndrome: Can result from paternal deletion of 15q11-q13 (70% of cases). Features include neonatal hypotonia, hyperphagia leading to obesity, intellectual disability, and behavioral problems.
  • Angelman Syndrome: Maternal deletion of the same 15q11-q13 region (70% of cases). Features include severe intellectual disability, seizures, ataxia, and a happy demeanor.
  • Potocki-Lupski Syndrome: Duplication of 17p11.2 (reciprocal of Smith-Magenis deletion) causes intellectual disability, autism, and congenital anomalies.
  • 22q11.2 Duplication Syndrome: Many features overlapping with the deletion syndrome but often milder, highlighting that both loss and gain of the same region can be pathogenic.

These examples underscore that CNVs can produce distinct syndromes depending on the genomic region altered, the genes involved, and the direction of the copy number change. The phenotypic spectrum of CNV-associated disorders is broad, influenced by incomplete penetrance and variable expressivity, which complicates genotype-phenotype correlations.

CNVs and Disease Risk

Beyond rare syndromic disorders, CNVs contribute to the risk of common complex diseases by altering gene dosage, disrupting regulatory networks, or unmasking recessive alleles. The contribution of CNVs to disease risk can be subtle, requiring large cohort studies for detection.

Neuropsychiatric Disorders

Autism Spectrum Disorder (ASD): Numerous studies have identified rare, large (often >250 kb) CNVs in individuals with ASD. Recurrent loci include 16p11.2 deletions and duplications, 7q11.23 duplications, 22q11.2 deletions, and 22q13.3 deletions (SHANK3). These CNVs are often de novo and have high penetrance. Patients with ASD may also carry multiple CNVs, suggesting an oligogenic or polygenic burden. The contribution of common small CNVs is more modest but collectively significant. NHGRI explains CNV prevalence in ASD.

Schizophrenia: Large-scale genome-wide association studies (GWAS) of CNVs have identified several risk loci, including deletions at 22q11.2 (OR ~20–30), 1q21.1, 3q29, 15q13.3, 16p11.2, and duplications at 7q11.23 and 16p13.1. These CNVs affect genes involved in synaptic plasticity, neuronal development, and immune function. The population attributable risk is low (each CNV explains <1% of cases), but for carriers, the risk of developing schizophrenia is substantially elevated. The 22q11.2 deletion is the strongest known genetic risk factor for schizophrenia, occurring in about 1–2% of patients.

Bipolar Disorder: While less studied, CNV burden is also elevated in bipolar disorder, with some overlap with schizophrenia-associated loci, suggesting shared genetic architecture. Recurrent duplications at 16p11.2 and deletions at 3q29 have been implicated.

Cancer

Somatic CNVs are hallmarks of cancer genomes. They can drive tumorigenesis by inducing oncogene amplification (e.g., ERBB2 in breast cancer, MYCN in neuroblastoma, EGFR in glioblastoma) or tumor suppressor gene deletion (CDKN2A/B on 9p21.3, PTEN on 10q23.31, TP53 on 17p13.1). Focal amplifications and deletions help define cancer subtypes and guide therapeutic decisions. For instance, ERBB2 (HER2) amplification is routinely tested in breast cancer to determine eligibility for trastuzumab. Germline CNVs can also predispose to cancer: for example, BRCA1 partial deletions cause hereditary breast-ovarian cancer; MSH2 deletions cause Lynch syndrome; and FHIT deletions are associated with various cancers. NCBI discusses CNVs in cancer genomics.

Cardiovascular and Metabolic Disorders

CNVs are increasingly linked to congenital heart disease (CHD). Deletions at 22q11.2, 1q21.1, 8p23.1, 16p13.11, and others are found in 3–10% of CHD patients. Large population-based studies have also shown that rare CNVs increase risk for hypertension, coronary artery disease, and type 2 diabetes. For example, a common deletion at 2q31.1 near the ZNF385B gene is associated with elevated fasting glucose and risk of diabetes. The mechanisms often involve altering the expression of genes critical for beta-cell function, insulin sensitivity, or vascular development.

Infectious Disease Susceptibility

Emerging evidence suggests CNVs influence host immune response to pathogens. The CCL3L1 gene copy number varies between individuals and populations; lower copy numbers are associated with increased susceptibility to HIV infection and progression to AIDS. Similarly, deletions affecting the complement component 4 (C4) genes modify risk for systemic lupus erythematosus and susceptibility to bacterial infections like meningococcus.

The study of CNVs in complex diseases is challenging due to their low frequency, requirement for large sample sizes, and the need for accurate CNV calling from array or sequencing data. Nonetheless, they represent an important piece of the genetic puzzle, often acting in concert with common SNPs and environmental factors.

Detection and Characterization of CNVs

Accurate detection of CNVs is essential for research and clinical diagnostics. Various technologies have been developed, each with strengths and limitations.

Chromosomal Microarray (CMA)

CMA, including array comparative genomic hybridization (aCGH) and SNP arrays, is the current first-tier diagnostic test for individuals with unexplained developmental delay, congenital anomalies, or autism spectrum disorder. aCGH compares patient DNA to a reference, identifying gains and losses across the genome at a resolution of 10–100 kb. SNP arrays can also detect CNVs and, in addition, identify regions of homozygosity (ROH) suggestive of uniparental disomy or consanguinity. CMA can detect both known pathogenic CNVs and novel variants of uncertain significance.

Next-Generation Sequencing (NGS)

Whole-genome sequencing (WGS) and whole-exome sequencing (WES) are increasingly used to identify CNVs, especially smaller or more complex alterations missed by arrays. Bioinformatics tools such as read-depth based methods (e.g., CNVnator, Control-FREEC) and split-read or paired-end mapping approaches can infer CNVs from sequencing data. WGS provides the most comprehensive view, including balanced rearrangements and CNVs in non-coding regions. However, computational challenges remain, particularly for detecting CNVs in repetitive regions and determining breakpoints accurately. Long-read sequencing (PacBio, Oxford Nanopore) is emerging as a powerful tool to resolve complex structural variants.

Targeted Methods

For known recurrent CNVs, targeted techniques like multiplex ligation-dependent probe amplification (MLPA) or quantitative polymerase chain reaction (qPCR) are cost-effective. FISH is used to visualize specific chromosomal rearrangements, especially in prenatal or cancer samples. However, these methods are limited to predefined regions and cannot discover new CNVs.

Clinical Interpretation and Databases

Interpreting CNV pathogenicity relies on population frequency, gene content, inheritance patterns, and functional evidence. Public databases such as DECIPHER (DECIPHER: database of genomic variants), ClinVar, and the Database of Genomic Variants (DGV) are essential resources. Guidelines from the American College of Medical Genetics and Genomics (ACMG) and Clinical Genome Resource (ClinGen) provide frameworks for classifying CNVs as pathogenic, likely pathogenic, benign, likely benign, or of uncertain significance. Recurrent CNVs with well-established phenotypes are easier to classify; novel or rare CNVs often require functional studies or segregation analysis.

Clinical Implications and Therapeutic Advances

The identification of CNVs has direct clinical applications in diagnosis, prognosis, and personalized medicine.

Prenatal and Preimplantation Genetic Testing

CMA is increasingly used in prenatal settings for detecting submicroscopic CNVs associated with neurodevelopmental disorders. Non-invasive prenatal testing (NIPT) using cell-free DNA can detect large CNVs (e.g., common aneuploidies and some microdeletions) with high accuracy. Preimplantation genetic testing (PGT) can screen for known familial CNV for couples at risk, enabling selection of unaffected embryos.

Pharmacogenomics

CNVs affecting drug metabolism genes have major consequences for drug efficacy and toxicity. For example, CYP2D6 copy number variations (including deletions, duplications, and multiplications) influence the metabolism of many antidepressants, antipsychotics, opioids, and tamoxifen. Poor metabolizers (deletion) require lower doses or alternative drugs; ultrarapid metabolizers (multiple copies) may not achieve therapeutic levels. Similar CNVs affect GSTM1, GSTT1, and UGT2B17, impacting detoxification and response to chemotherapies. PharmGKB details CYP2D6 CNV dosing guidelines.

Cancer Targeted Therapy

Detection of CNVs is integral to precision oncology. HER2 amplification in breast cancer, EGFR amplification in lung cancer, and MET amplification in gastric cancer predict response to targeted kinase inhibitors. Loss of PTEN or CDKN2A may predict resistance to certain therapies or sensitivity to CDK4/6 inhibitors. Liquid biopsies using circulating tumor DNA can now detect CNVs in plasma, enabling non-invasive monitoring of tumor evolution and resistance.

Gene Therapy and CRISPR

In principle, CNV-associated disorders caused by haploinsufficiency might be treated by gene augmentation (e.g., delivering a functional copy of the missing gene via AAV vectors) or by activating the remaining allele. For gain-of-function duplications, strategies such as RNA interference or CRISPR-mediated silencing are being explored. However, challenges include delivering genes to the right tissues, achieving appropriate expression levels, and avoiding off-target effects. For large CNVs that delete entire gene clusters, replacement remains difficult; future approaches may involve synthetic chromosomes or targeted gene editing in specific cells.

Future Directions

Ongoing research continues to refine our understanding of CNVs at both molecular and clinical levels. The integration of long-read sequencing into clinical workflows will enable detection of complex structural variants that are currently missed or miscalled. Single-cell sequencing allows the study of somatic CNVs within tissues, revealing clonal heterogeneity in cancer and mosaicism in neurodevelopmental disorders. Large-scale biobank studies (e.g., UK Biobank, All of Us) are leveraging array and sequencing data to discover new CNV-disease associations with greater statistical power. Functional genomics tools, such as CRISPR-Cas9 screens and 3D chromatin mapping, are illuminating how CNVs alter gene regulation and cellular phenotypes. Ultimately, the goal is to translate these discoveries into improved diagnostics, risk prediction, and therapeutic interventions across the spectrum of human disease.

Conclusion

Copy number variations represent a fundamental layer of human genetic variation, with profound implications for both rare disorders and common diseases. From well-known microdeletion syndromes to the complex genetic architecture of autism, schizophrenia, and cancer, CNVs influence disease susceptibility through dosage-sensitive genes and regulatory elements. Technological advances in detection have made CNV analysis a routine part of clinical genetics, enabling accurate diagnoses and guiding treatment decisions. As research uncovers the full extent of CNV functional impact, and as therapeutic strategies advance, the translation of CNV knowledge into clinical practice promises to improve patient outcomes. Understanding CNVs is not merely an academic exercise—it is a critical component of precision medicine and a pathway to better health.