SNP Genotyping Test: DNA Variants, Risk Markers, and Results

SNP genotyping tests identify selected single-nucleotide polymorphisms—positions in DNA where people commonly differ by one base. A test may examine one medically important variant, hundreds of pharmacogenetic markers, or hundreds of thousands of sites on a genotyping array. The resulting genotype can provide useful information about ancestry, traits, medication response, carrier status, or statistical susceptibility to disease. It usually does not read every base of a gene and should not be confused with comprehensive sequencing.

The clinical meaning of a SNP depends on why it was selected and how strong the supporting evidence is. Some genotypes directly alter a protein or help define a validated drug-metabolism phenotype. Others are only markers located near the true biological cause. A risk-associated allele may modestly shift probability without determining whether disease will occur. Results are also affected by ancestry, environmental exposures, age, family history, and the population used to build the risk estimate.

• SNP genotyping checks predetermined DNA positions rather than searching the entire genome for all variants.
• A genotype may be reported as two base letters, an allele count, a haplotype, or a calculated risk score.
• Most common risk markers change probability; they do not diagnose a disease.
• Array tests can be accurate for common intended markers but unreliable for some rare variants.
• Health-related findings may require confirmation in a clinical laboratory before care is changed.

Table of Contents

• From Single-Base Change to Genotype
• Different Purposes of SNP Testing
• How Genotyping Assays Work
• Reading Alleles, Haplotypes, and Risk Scores
• Clinical Meaning of Risk Markers
• Accuracy, Ancestry, and Major Limitations
• Responsible Follow-Up After a Result

From Single-Base Change to Genotype

A single-nucleotide polymorphism, or SNP, is a genomic position at which a single DNA base varies among people. The term is commonly used for variants that occur relatively often in a population, although laboratories and reports sometimes use “SNP” more loosely for any single-base variant. In modern clinical language, “single-nucleotide variant” is the broader term.

Humans generally carry two copies of each autosomal position, one inherited from each parent. If a site can contain either A or G, a person’s genotype might be AA, AG, or GG. AA and GG are homozygous genotypes; AG is heterozygous. The letters may be reported on the forward genomic strand, the reverse strand, or according to a named allele system. Strand orientation matters because A pairs with T and C pairs with G. Raw data from different companies can appear inconsistent when they use opposite orientations even though the underlying genotype is the same.

Many SNPs have no known health effect. They can still be useful as landmarks because nearby DNA tends to be inherited in blocks. A marker may be statistically associated with a disease because it is in linkage disequilibrium with another causal variant. The marker itself may not change a protein or gene function.

Genotyping differs from sequencing. A genotyping assay asks, “Which of the expected alleles is present at this known position?” Sequencing asks, “What is the order of bases across this region, including unexpected changes?” A microarray may test hundreds of thousands of SNPs efficiently but will usually miss a rare pathogenic variant that is not represented on the chip. A whole-genome sequencing test surveys far more positions and variant types, although it introduces different analytic and interpretation challenges.

Some SNPs have a direct functional effect. A base change may alter an amino acid, influence RNA splicing, affect gene expression, or help define an enzyme haplotype. Others serve only as proxies. A report should distinguish a tested causal variant from a tag SNP and should avoid implying that every association explains the biological mechanism.

SNP genotyping can also be used for identity, sample tracking, kinship, ancestry inference, and research. These applications use patterns across many sites rather than treating one SNP as a diagnosis. The same raw genotype may be interpreted differently depending on the validated purpose of the assay.

Different Purposes of SNP Testing

The phrase “SNP test” covers several very different products. Understanding the test’s intended use is essential before interpreting any result.

Targeted medical genotyping

A targeted test examines one or a small number of established variants. It may be ordered to detect a common disease-associated allele, determine whether a person carries a founder variant, or support a diagnosis in a defined clinical setting. Targeted testing can be fast and inexpensive, but a negative result excludes only the specific variants tested.

For example, a product may test selected variants associated with hereditary cancer. If none are found, the person could still carry one of many other pathogenic variants in the same genes. Targeted genotyping should not replace full gene analysis when personal or family history meets criteria for comprehensive testing.

Pharmacogenetic testing

Pharmacogenetic assays examine variants that affect drug metabolism, transport, targets, or adverse-reaction risk. Several SNPs may be combined into a star allele or haplotype, which is then translated into a phenotype such as poor, intermediate, normal, rapid, or ultrarapid metabolizer.

This translation is not always simple. Genes such as CYP2D6 can have deletions, duplications, hybrid alleles, and complex haplotypes that a basic SNP panel may not resolve. A result that checks only common variants can assign the wrong phenotype in people with an untested allele. Clinical use should rely on a validated assay and current prescribing guidance, not a raw SNP list.

Disease susceptibility markers

Genome-wide association studies compare SNP frequencies in people with and without a condition. They identify variants associated with a change in relative risk. Most individual common SNPs have small effects. A person can carry a risk allele and never develop the disease, while someone without it can still become affected.

Risk markers are most useful when the association is replicated, the effect estimate is relevant to the person’s ancestry, and the result adds information beyond standard risk factors. A marker with an odds ratio of 1.2 does not mean a 20% absolute chance of disease. It means the odds differ relative to a comparison group, and the absolute effect depends on baseline risk.

Polygenic scores

A polygenic risk score combines weighted effects from many SNPs. Some scores use dozens of variants; others use millions through direct genotypes and imputation. The score usually places a person within a distribution, such as a percentile compared with a reference population.

Polygenic scores are being evaluated for cardiovascular disease, diabetes, cancers, and other common conditions. Clinical readiness varies by condition and setting. A score is not a universal measure of genetic health, and one model cannot automatically be transferred across populations or laboratories.

Ancestry, traits, and wellness testing

Consumer tests often use dense SNP arrays to estimate genetic ancestry, relatives, physical traits, and wellness characteristics. Ancestry percentages depend on the company’s reference groups and algorithms and can change when databases are updated. They describe genetic similarity to sampled populations, not culture, nationality, or personal identity.

Trait predictions may be entertaining or educational but often have limited medical significance. Health reports offered alongside them must be evaluated separately. Regulatory review, clinical validation, and scope differ among products.

Research and biobanking

SNP arrays are widely used in research to study population structure, identify disease associations, perform quality control, and impute untyped variants. A research result may not meet clinical laboratory standards and may lack individual confirmation or interpretation. Participants should not use research-only findings for medical decisions unless they are returned through an appropriate clinical process.

How Genotyping Assays Work

SNP genotyping can be performed with several technologies. All depend on recognizing the DNA base at a predetermined site, but they differ in scale and ability to handle complex regions.

Allele-specific PCR

Allele-specific PCR uses primers or probes designed to distinguish one base from another. A fluorescent signal indicates which allele amplified. This approach is efficient for a small set of variants and is commonly used in targeted medical and pharmacogenetic assays.

The test needs positive and negative controls because poor DNA or a nearby unrecognized variant can interfere with primer binding. An absent signal should not automatically be interpreted as a rare genotype; it may represent assay failure.

SNP microarrays

A microarray contains probes for hundreds of thousands or millions of selected sites. DNA is fragmented, labeled, and allowed to bind to matching probes. Signal intensity and pattern are converted into genotype calls using clustering algorithms.

Arrays are efficient and generally accurate for common variants included in the design. They can also support ancestry analysis, polygenic scoring, and detection of some large copy-number changes or long stretches of homozygosity. Their performance is weaker for rare variants, small insertions and deletions, complex structural variants, and sites with poor probe behavior.

A chip does not directly type every variant later shown in a report. Some positions may be imputed—statistically inferred from nearby genotypes and a reference panel. Imputation can be useful at population scale but is probabilistic. Accuracy depends on the array, marker density, reference population, local genomic structure, and the individual’s ancestry.

Sequencing-based genotyping

Targeted sequencing or broader NGS can also call selected SNPs. Sequencing may identify nearby unexpected variants and can provide read-depth evidence, but it still must be validated for the intended purpose. A report that extracts several SNPs from exome or genome data may not have uniform coverage at every pharmacogenetic or ancestry marker.

Mass spectrometry and other platforms

Some laboratories use primer extension with mass spectrometry, bead-based assays, restriction analysis, or digital methods. The platform matters less than whether the exact variants, specimen types, and interpretation rules were analytically validated.

The specimen is usually blood or saliva. Cheek swabs may also be accepted. DNA quality affects call rate. Low concentration, contamination, degradation, or poor saliva collection can produce missing genotypes. Consumer laboratories may request a repeat sample when too many sites fail quality control.

The laboratory’s quality process should assess sample identity, overall call rate, duplicate concordance, control performance, and confidence at each site. For a result with major health consequences, direct visual or orthogonal review may be appropriate rather than relying entirely on an automated array call.

Reading Alleles, Haplotypes, and Risk Scores

A SNP report may appear simple, but several layers separate a base call from a medical recommendation.

Reference SNP identifiers

Many variants are labeled with an “rs” number, such as rs followed by digits. This identifier points to a database record, not to a clinical interpretation. Database records can merge, change, or contain multiple genomic placements. A clinically usable report should also specify genome build, gene, alleles, and orientation when relevant.

Risk and non-risk alleles

A report may label one allele “risk increasing,” “protective,” or “typical.” These terms are relative to a study population. They do not mean that one allele is universally harmful. The effect may differ by ancestry, sex, age, exposure, or disease subtype. A protective association rarely reduces risk to zero.

Haplotype and diplotype

A haplotype is a set of variants inherited together on one chromosome. A diplotype is the pair of haplotypes carried by a person. Pharmacogenetic reports often translate diplotypes into predicted enzyme activity.

Standard genotyping reveals which alleles are present but may not show which alleles lie on the same chromosome. Statistical phasing estimates this arrangement. Family data or long-read methods may be needed when phase materially changes interpretation.

Relative versus absolute risk

Relative risk compares groups. Absolute risk estimates the probability that a person will develop a condition over a specified period. A relative increase can sound dramatic while producing a small absolute change when baseline risk is low.

For example, doubling a 1% baseline risk produces an approximate 2% risk, not certainty. Conversely, a modest relative increase can matter when baseline risk is high. A useful report should provide context, time frame, comparison population, and major nongenetic factors.

Percentiles and polygenic distributions

A person in the 90th percentile for a score has a value higher than about 90% of the chosen reference group. It does not mean a 90% chance of disease. Translating a percentile into absolute risk requires validated incidence data and calibration for ancestry, age, sex, and clinical factors.

Different companies can produce different scores from the same DNA because they use different variants, weights, genome builds, imputation panels, and reference populations. Scores should not be compared across systems unless specifically harmonized.

Positive, negative, and no-call results

For a targeted variant, “positive” means the tested allele was detected. “Negative” means it was not detected, not that the gene or disease risk is absent. A no-call means the assay could not assign a reliable genotype. Recollection or confirmation may be necessary.

A raw data file is not a clinical report. Third-party websites may interpret thousands of SNPs without verifying orientation, call quality, current evidence, or clinical validity. Apparent rare pathogenic findings from consumer arrays have a meaningful risk of being false and should be confirmed before they influence care.

Clinical Meaning of Risk Markers

A risk marker has clinical value only when it improves a decision. Statistical association alone is not enough. The result should add information beyond age, blood pressure, cholesterol, family history, smoking, imaging, or other established predictors and should lead to an action supported by evidence.

Three concepts help evaluate a marker:

• Analytic validity: Does the test accurately identify the genotype?
• Clinical validity: Is the genotype reliably associated with the stated condition or response?
• Clinical utility: Does using the result improve health decisions or outcomes?

A test can have excellent analytic validity but weak clinical utility. An array may call a common SNP correctly, yet the SNP’s effect may be too small or uncertain to change screening.

For monogenic disorders, a pathogenic variant can have a large effect and may confirm a diagnosis. Most SNP susceptibility markers behave differently. They are one contribution among many and should not be described with deterministic language such as “you will develop” or “you are safe.”

Polygenic risk scores may eventually help identify people who benefit from earlier screening or more intensive prevention. Their use should be tied to a defined clinical pathway. Simply telling someone that they are in a high percentile without offering a validated management plan can create anxiety without benefit.

Risk estimates must be calibrated. A score developed mainly in one ancestry group may rank or estimate risk less accurately in another group because allele frequencies, linkage patterns, and environmental contexts differ. This is both a scientific and an equity concern. Laboratories should disclose the populations used for development and validation.

Pharmacogenetic results are somewhat different because they may guide selection or dosing of a specific medication. Even then, genotype is not the only factor. Kidney and liver function, age, drug interactions, adherence, and the treatment indication can change the recommendation. A clinician should use an established guideline and verify that the assay captured the relevant alleles.

Disease screening should continue according to personal and family history unless a qualified clinician recommends otherwise. A low genetic score does not cancel strong clinical risk factors. A high score is not a diagnosis and may not justify imaging, medication, or surgery outside professional guidance.

Accuracy, Ancestry, and Major Limitations

Genotyping accuracy is site specific. A company may report a very high overall call rate while a particular rare or medically important variant is miscalled. Common SNPs used to form large clusters generally perform better than rare alleles represented by very few reference samples.

Rare pathogenic variants are a known weak point for some SNP chips. Automated algorithms can place an unusual signal into the wrong genotype cluster. This is especially concerning when raw consumer data are searched for rare disease variants that the product was not designed to report. Clinical confirmation with targeted sequencing is essential.

Other limitations include:

• Limited content: Only variants selected for the assay are directly tested.
• Version differences: Newer chip versions may add or remove markers, making results unequal across customers.
• Imputation uncertainty: An inferred genotype is not the same as a directly measured one.
• Structural complexity: Copy-number changes, repeats, pseudogenes, and gene conversions may not be resolved.
• Phase uncertainty: Separate alleles may be detected without knowing whether they lie on the same chromosome.
• Population bias: Association strength and score calibration may not transfer across ancestries.
• Changing science: A reported association may weaken, strengthen, or disappear as larger studies are published.
• Phenotype oversimplification: Broad labels such as “heart disease” can combine biologically different conditions.

Ancestry estimates have their own limitations. They depend on who is represented in the reference database and how populations are defined. Nearby regions are assigned probabilistically, and small percentages may reflect noise or shared ancient ancestry. Identical twins can receive slightly different results because of technical and algorithmic variation, while siblings normally inherit different mixtures from their parents.

Privacy deserves attention. Dense SNP data are identifying and reveal information about biological relatives. Database matching can uncover unexpected parentage or previously unknown relatives. Policies for data storage, research use, law-enforcement access, account deletion, and sample destruction vary. Consumers should read current terms rather than assuming that “de-identified” data cannot be connected to them.

A health-related genotype may also have insurance or discrimination implications depending on jurisdiction. Legal protections vary and may not cover life, disability, or long-term-care insurance. A genetic counselor can discuss these issues before confirmatory testing.

Responsible Follow-Up After a Result

The appropriate response depends on whether the result comes from a clinical assay, a direct-to-consumer report, or an unofficial interpretation of raw data.

For a medically important result, first verify exactly what was measured. Determine whether the variant was directly genotyped or imputed, whether the laboratory is clinically accredited, and whether the report recommends confirmation. Do not change medication, screening, or surgery based only on a third-party raw-data interpretation.

Confirmation is especially important when:

• the reported variant is rare or pathogenic;
• the finding conflicts with family history or prior testing;
• the call comes from an array rather than sequencing;
• the decision could lead to medication changes, invasive screening, or preventive surgery;
• the test examined only selected founder variants;
• the result has implications for relatives or pregnancy.

A clinician may order a focused targeted variant test or comprehensive gene analysis. Confirmation should use a new specimen when sample identity is a concern.

For a common risk marker or polygenic score, ask for the absolute risk, reference population, and evidence that using the result improves outcomes. Place it beside standard clinical factors. Healthy actions such as avoiding tobacco, exercising, maintaining blood pressure, and attending recommended screening usually remain important regardless of genotype.

For a pharmacogenetic result, a pharmacist or prescribing clinician should review the gene-drug guideline, current medication, dose, interactions, and assay coverage. The genotype may be relevant for future prescriptions, so a verified result should be stored in the medical record in a form that can trigger decision support.

For ancestry or relative matching, users should prepare for unexpected discoveries and consider whether relatives consent to the information being revealed. Small ancestry updates do not require medical follow-up.

A no-call or contradictory result should not be forced into an interpretation. Repeating collection or using another validated method is preferable. When several companies disagree, differences in tested markers and algorithms are often more likely than a change in the person’s DNA.

SNP genotyping is powerful when its scope is respected. It can provide reliable answers at selected sites and can summarize patterns across the genome, but it cannot substitute for comprehensive diagnostic sequencing, clinical risk assessment, or professional interpretation.

References

• Clinical utility and implementation of polygenic risk scores for cardiovascular disease: a clinical consensus statement of the ESC Council on Cardiovascular Genomics, the European Association of Preventive Cardiology, the European Association of Cardiovascular Imaging, and the European Heart Rhythm Association — 2025 Consensus Statement.
• Comparing commercial pharmacogenetic testing results and recommendations for antidepressants — 2024 Study.
• Human Genetics Society of Australasia Position Statement: Use of polygenic scores in clinical practice and population health — 2023 Position Statement.
• Clinical utility of polygenic scores for common diseases: A practice resource from the National Society of Genetic Counselors — 2023 Practice Resource.
• Extent to which array genotyping and imputation with large reference panels approximate deep whole-genome sequencing — 2022 Study.
• Use of SNP chips to detect rare pathogenic variants: retrospective, population based diagnostic evaluation — 2021 Study.

Disclaimer

This article is for education and does not replace medical, pharmacologic, or genetic advice. SNP tests differ in content, validation, population relevance, and regulatory status. Clinically important consumer or research findings should be confirmed and interpreted by qualified professionals before health decisions are made.