An SNV, or single nucleotide variant, is a change at a single position in a DNA sequence where one nucleotide letter (A, T, C, or G) is swapped for another. It is the smallest and most common type of genetic variation in the human genome. On average, any person’s genome contains roughly 5 million SNVs compared to the human reference sequence, accounting for much of the 0.4% that makes your DNA unique.
SNV vs. SNP: What’s the Difference?
You’ll often see “SNV” and “SNP” used almost interchangeably, but they mean slightly different things. SNV is the broader term: any single-letter change at any frequency, including one that exists in only a single person. An SNP (single nucleotide polymorphism) is an SNV that has become common enough in a population to cross a frequency threshold, typically a minor allele frequency of at least 1% to 2%. In large validation studies, researchers have used a cutoff of 2% to classify a variant as a confirmed, polymorphic SNP.
In practical terms, if a variant is brand new (appearing for the first time in one individual), it’s an SNV but not an SNP. If millions of people carry it, it qualifies as both. The distinction matters in clinical genetics because rare SNVs are more likely to be medically significant, while common SNPs often have little or no impact on health.
How SNVs Arise
Every time a cell divides, its entire genome has to be copied. The molecular machinery that handles this job is remarkably accurate, but not perfect. Errors during DNA replication are the primary source of new SNVs. The enzymes that copy DNA occasionally insert the wrong nucleotide, and while most mistakes get caught and corrected by proofreading systems, some slip through.
Other sources include chemical damage to DNA bases (such as spontaneous deamination, where a C converts to a U and is then read as a T) and exposure to external mutagens like ultraviolet radiation or tobacco smoke. DNA repair processes themselves can also introduce errors. Research has shown that certain repair pathways use lower-fidelity copying enzymes, which increases the local rate of point mutations in the regions they touch.
Whole-genome sequencing studies across nearly 100 families have measured the average rate of new (de novo) SNVs in the human germline at about 1.16 × 10⁻⁸ mutations per base pair per generation. That translates to roughly 70 brand-new single-letter changes in every baby’s genome that neither parent carried. This rate is consistent across studies of diverse populations, including Icelandic trios and Hutterite pedigrees.
Types of SNVs and Their Effects
Where an SNV lands in the genome determines how much it matters. Most of your DNA sits outside of genes, so many SNVs fall in regions with no obvious function and have no detectable effect. When an SNV does occur inside a gene’s protein-coding region, its consequences depend on how the change alters the genetic instructions.
- Synonymous (silent) SNVs change a DNA letter but don’t change the amino acid the gene codes for, because multiple three-letter codes can specify the same amino acid. These were long considered harmless, but that view is outdated. A synonymous change can swap in a “rare” codon that the cell’s protein-building machinery reads more slowly. This altered speed can disrupt how a protein folds as it’s being made, sometimes producing a misfolded protein with reduced or altered activity.
- Missense SNVs substitute one amino acid for another in the resulting protein. The impact ranges from negligible (if the two amino acids are chemically similar) to severe (if the swap occurs at a critical spot in the protein’s structure). Sickle cell disease, for example, results from a single missense SNV in the hemoglobin gene.
- Nonsense SNVs convert a normal amino acid code into a premature stop signal, cutting the protein short. This usually destroys the protein’s function entirely.
SNVs outside coding regions can still have consequences. Changes in regulatory regions may turn a gene’s activity up or down, while SNVs at the boundaries between coding and non-coding segments (splice sites) can cause entire sections of a gene to be skipped or incorrectly included.
How SNVs Are Classified Clinically
When genetic testing identifies an SNV, it gets sorted into one of five categories established by the American College of Medical Genetics and Genomics: pathogenic, likely pathogenic, variant of uncertain significance (VUS), likely benign, or benign. Each classification is tied to a specific condition and inheritance pattern, so the same variant might be pathogenic for one disease and irrelevant to another.
The classification draws on multiple lines of evidence. If a missense SNV produces the same amino acid change as a previously confirmed disease-causing variant (even through a different nucleotide swap), it can generally be considered pathogenic as well. Population frequency plays a role too: a variant found commonly in healthy people is unlikely to cause a serious genetic disorder. Family studies, functional laboratory experiments, and computational predictions of protein damage all feed into the final call.
A VUS classification is common and often frustrating for patients. It means the variant was detected but there isn’t enough evidence yet to say whether it’s harmful or harmless. As more genomes are sequenced and more functional data accumulate, many VUS results get reclassified over time.
How SNVs Are Detected
Next-generation sequencing (NGS) is the standard technology for identifying SNVs. A DNA sample is broken into millions of short fragments, each fragment is read independently, and software assembles the reads and compares them to a reference genome to spot single-letter differences.
Accuracy depends heavily on how many times each position in the genome gets read, a measure called coverage depth. For inherited (constitutional) variants, where roughly half the DNA molecules carry the change, moderate coverage is sufficient. Detecting SNVs in tumor samples is harder because cancer tissue is a mix of normal and mutant cells, and a variant might be present in only a small fraction of the DNA. This requires much deeper sequencing and specialized analysis software calibrated with control samples to distinguish real low-frequency variants from sequencing errors.
Other factors that influence detection quality include how the DNA was prepared (samples preserved in formalin, common in cancer biopsies, tend to have more artifacts) and whether the sequencing targets specific gene panels or the whole genome.
How SNVs Are Written Down
Geneticists use a standardized notation maintained by the Human Genome Variation Society (HGVS) so that any variant can be described unambiguously. An SNV is written with a prefix indicating the reference type (“g.” for genomic position, “c.” for coding DNA, “p.” for protein), followed by the position number and the base change. For example, c.34G>T means the guanine at position 34 in the coding sequence was replaced by thymine. The corresponding protein-level notation, p.Val12Leu, tells you that valine at position 12 was replaced by leucine.
Clinical genetic reports typically present variants in a table that includes both DNA-level and protein-level names, along with the gene, the associated condition, the inheritance pattern, and the variant’s classification. This layered naming system ensures that a variant identified in one lab can be looked up, compared, and reinterpreted anywhere in the world.