SNPs (pronounced “snips”) are the most common type of genetic variation among people. Short for single nucleotide polymorphisms, they are spots in your DNA where a single “letter” of the genetic code differs from person to person. On average, a single-letter difference appears between any two people’s genomes once every 1,300 DNA letters or so, which is why humans are often said to be 99.9% genetically identical.
How SNPs Differ From Random Mutations
Your DNA is written in a four-letter alphabet: A, T, C, and G. A SNP is a position where one of those letters has been swapped for another, and that swap shows up in at least 1% of the human population. That 1% threshold is what separates a SNP from a rare mutation. If a variant is carried by fewer than 1 in 100 people, researchers classify it as a rare variant rather than a polymorphism.
Most SNPs are ancient. Common variants tend to be thousands of years old, having persisted through generations because they were either neutral or mildly beneficial. The main public database for these variants, called dbSNP, now catalogs over 1.1 billion unique reference SNPs collected from roughly 3,900 labs across 81 countries. That number reflects how widespread and well-documented this type of variation has become.
Where SNPs Sit in Your DNA
A SNP can land almost anywhere in the genome, and its location determines whether it matters biologically. Most SNPs fall in stretches of DNA that don’t code for proteins, and many of those have no detectable effect at all. These are considered neutral.
Functional SNPs are the ones that change something. A SNP in a gene’s promoter region (the switch that turns a gene on or off) can raise or lower how much protein that gene produces. A SNP inside a protein-coding region can swap one amino acid for another, potentially altering the shape or function of the resulting protein. SNPs in the non-coding sections between genes can still influence how DNA is read by affecting a process called splicing, where the cell edits genetic instructions before building a protein. Some SNPs even change the behavior of tiny regulatory molecules called microRNAs, which act as volume knobs for other genes.
The short version: location is everything. Two SNPs sitting just a few hundred letters apart can have wildly different consequences, or no consequence at all.
SNPs and Disease Risk
Researchers use large-scale studies called genome-wide association studies (GWAS) to scan the DNA of thousands or millions of people, looking for SNPs that show up more often in those with a particular disease. This approach has linked specific SNPs to conditions ranging from heart disease to diabetes to Alzheimer’s.
Alzheimer’s provides one of the clearest examples. The APOE gene has three common versions, labeled ε2, ε3, and ε4. Carrying the ε4 version increases your risk for Alzheimer’s and tends to push the age of onset earlier. The ε2 version, by contrast, appears protective. Beyond APOE, GWAS have identified SNPs in several other genes (SORL1, PICALM, CLU, and CR1) that also raise Alzheimer’s risk, likely by influencing the buildup of amyloid protein in the brain.
For most common diseases, though, no single SNP tells the whole story. Heart disease, type 2 diabetes, and depression are influenced by hundreds or even thousands of SNPs, each contributing a small nudge in risk. To capture this complexity, scientists calculate what’s called a polygenic risk score. It works by adding up the effects of all relevant SNPs in a person’s genome, weighted by how strongly each one is associated with the disease. The result is a single number estimating your overall genetic predisposition.
How SNPs Affect Medication Response
One of the most practical applications of SNP research is pharmacogenomics, which uses genetic variation to predict how you’ll respond to certain drugs. Two well-studied examples involve common blood-thinning and anti-clotting medications.
The blood thinner warfarin is processed by an enzyme whose efficiency varies depending on which SNPs you carry. People with certain variants in the CYP2C9 gene clear the drug 40 to 70% more slowly, which raises their bleeding risk, especially in the first weeks of treatment. Another SNP in a gene called VKORC1 also shifts the dose a person needs. Warfarin’s FDA label now includes a dosing table based on these genetic variants.
A similar story plays out with clopidogrel, a drug used to prevent blood clots after heart attacks or stent placement. A common SNP in the CYP2C19 gene produces a version of the enzyme that can’t activate the drug properly. People who carry two copies of this loss-of-function variant get little benefit from the standard dose, and the FDA added a boxed warning recommending alternative therapy for these patients. On the other end of the spectrum, a different CYP2C19 variant makes the enzyme overactive, leading to stronger drug effects and higher bleeding risk.
SNPs in Ancestry and Consumer Testing
Because common SNPs accumulated over thousands of years of human migration, they serve as useful markers for tracing where your ancestors lived. Africa has the highest number of region-specific SNPs, which supports the “Out of Africa” model of human origins. The Americas, isolated from major Eurasian migration routes for roughly 10,000 years, developed their own distinctive set of variants. By comparing your SNP profile against reference populations from different continents, testing companies can estimate the geographic origins of your ancestors.
Consumer genetic tests from companies like 23andMe and AncestryDNA typically analyze hundreds of thousands of SNPs using a technology called a microarray, which is essentially a chip that checks for known variants at specific positions. These kits report on ancestry composition, but they also flag SNPs associated with traits like eye color, male-pattern baldness, bitter taste perception, and drug metabolism. The accuracy of trait predictions varies. Eye color, which is influenced by a relatively small number of SNPs, can be predicted fairly reliably. Complex traits like athletic ability or intelligence involve so many genetic and environmental factors that current SNP-based predictions remain rough estimates.
How SNPs Are Detected
There are two main approaches to identifying SNPs. Microarrays (sometimes called SNP chips) check for known variants at predetermined positions across the genome. They’re relatively affordable and produce high-quality data, which is why they dominate consumer testing and many clinical applications. The tradeoff is that they only detect variants the chip was designed to look for.
Sequencing-based methods read the DNA more broadly, capturing both known and previously undiscovered variants. This approach generates far more data points but comes with higher costs and more missing data that requires statistical correction. For large research projects, both methods identify the same broad genetic patterns in a population. Sequencing is increasingly favored in research settings for its ability to pick up rare variants that microarrays would miss, while arrays remain the workhorse for standardized, large-scale genotyping.
What a Single SNP Can and Can’t Tell You
It’s tempting to think of SNPs as simple on/off switches, but most work more like dimmers. A single SNP rarely determines whether you will develop a disease or respond to a drug. Instead, it shifts your probability by a small amount. The exceptions, like the APOE ε4 variant for Alzheimer’s, are notable precisely because they carry outsized influence.
For the vast majority of traits and conditions, what matters is the cumulative picture: thousands of SNPs interacting with each other and with your environment, diet, sleep, stress, and exposures over a lifetime. SNPs give you one layer of information about your biology. They’re a powerful layer, and an increasingly useful one, but they’re never the full story.