Genotyping and sequencing both help researchers describe DNA, but the words refer to different aspects of measurement. Genotyping determines which alleles are present at particular genomic locations. Sequencing determines the order of DNA bases in the material examined. Sequencing data can therefore be used to determine genotypes. In everyday comparisons, however, genotyping often means an assay that tests a predefined set of genetic markers, while sequencing means reading stretches of DNA.
That distinction matters more than which method sounds more comprehensive. The useful question is whether an approach measures the information needed for a particular purpose. Our overview of genotyping explains the underlying terminology; this guide compares how the methods collect evidence and where their limits arise.
What targeted genotyping measures
Imagine a hypothetical position at which two possible alleles are A and G. A genotyping assay designed for that position may return A/A, A/G, or G/G for a typical autosomal location. It may also return no call if the evidence does not meet its quality requirements. The result describes the tested position, not every base around it.
A genotyping array extends this approach across many selected markers. It uses nucleic acid probes on a solid surface, with detectable signals indicating how sample material interacts with those probes. The chip design determines which markers can be examined directly. Researchers also use other laboratory methods to genotype selected variants; genotyping is broader than a single kind of chip.
A list of markers distributed across the genome can support a genome-wide analysis without reading the entire genome. Distribution and completeness are different properties. If a region contains a variant outside the assay's selected targets, a standard marker result does not reveal whether that untested variant is present.
A missing call also has a specific meaning: the workflow did not assign a usable genotype. It should not be treated as evidence that a reference allele is present. When comparing datasets, keeping missing observations separate from reported genotypes prevents an empty field from being mistaken for a biological finding.
What sequencing measures
Sequencing produces observations of base order. Depending on the technology and experiment, those observations may be short or long stretches called reads. A common analysis aligns reads to a reference sequence, evaluates the supporting evidence, and identifies differences through variant calling. Other approaches assemble sequences from overlapping observations.
Sequencing can identify differences that were not individually selected as assay targets beforehand. Its ability to do so still depends on which regions were included, how well they were read, and what kinds of variation the analysis can detect. Reading DNA, recognizing a difference, and explaining its significance are separate steps.
Consider an invented ten-base region with an unexpected substitution at its seventh base. An assay that tests only the third base could report its intended genotype correctly while providing no direct information about the seventh. Sequencing that adequately covers the full region may detect the substitution. This example illustrates the difference in scope; it does not establish that every sequencing workflow finds every possible change.
Sequencing has several possible scopes
A sequencing test can examine one region, a defined set of genes, the exome, or the genome. The method name alone does not tell you which of these was used. A small target sequenced thoroughly can answer a different question from a broad survey with less detailed observation at each position.
- Targeted sequencing: examines specified regions. A panel generally includes a defined collection of genes or other selected targets.
- Exome sequencing: concentrates on protein-coding portions of the genome and associated targeted regions, with the exact design varying by assay.
- Genome sequencing: aims to collect sequence information across the genome, including coding and noncoding regions.
MedlinePlus provides an accessible comparison of whole exome and whole genome sequencing. When reading a specific test description, check the target definition and the reported limitations. A broad collection strategy does not ensure that every part is equally observable or that every collected result will appear in the final report.
Coverage describes evidence, not certainty
Depth of coverage refers to how many sequencing observations cover a position. Repeated observations can help distinguish a genuine sequence difference from an error. But an average depth is an average: a hypothetical experiment averaging 30 observations per position can contain locations with substantially more or fewer observations.
Coverage is also used to describe the breadth of a target that has usable data, so the definition beside a number matters. Ask whether a figure describes average depth, the fraction of bases passing a threshold, or another measure. These quantities address different gaps in the evidence.
Some genomic regions are difficult to analyze because their sequences repeat or closely resemble other regions. Reads may be hard to place uniquely, and different technologies handle these challenges differently. Benchmark studies evaluate performance for particular variant types and genomic contexts. A single accuracy claim can conceal that variation unless its denominator and scope are stated.
Imputation estimates information that was not directly measured
Imputation uses patterns in observed genotypes and reference data to estimate genotypes at additional positions. Nearby alleles often occur together in recognizable combinations, allowing statistical methods to make informed estimates. This can expand the usefulness of marker data in research, particularly when comparing studies that measured different sets of variants.
An imputed genotype is an estimate with uncertainty. Its reliability depends on the available observations, the variant, the method, and how well the reference data represent the relevant genetic variation. A large imputed marker count should therefore be distinguished from the count directly assayed. More rows in a downloadable file do not necessarily mean more independent laboratory measurements.
A useful methods description identifies which results were observed, which were inferred, and what quality measures accompany the inference. The same principle applies when comparing a study's conclusions across populations: performance demonstrated in one dataset does not automatically establish identical performance in every other dataset.
Variant detection depends on the whole workflow
Small substitutions, missing segments, extra copies, and repeat expansions pose different measurement problems. A workflow optimized for one class may need additional methods or analysis to characterize another. Some arrays can assess copy-number changes, while sequencing workflows differ in their ability to detect structural variation. Treat each claimed capability as something to verify in the assay description.
The sample and analysis also matter. Poor material, mixed samples, filtering choices, and technical artifacts can affect results. Quality control is therefore part of both genotyping and sequencing. Comparing the names of two technologies cannot substitute for comparing how the actual experiments were designed and validated.
A measured variant still needs interpretation
A laboratory can correctly identify a sequence difference without establishing a clear connection to a trait. Analytical validity concerns detection accuracy; clinical validity concerns the relationship between the finding and a health outcome. These are different questions, and strong evidence for one does not automatically establish the other.
Likewise, the final report may include only a subset of the measured data. The report's purpose, analysis filters, and evidence criteria shape what appears. Our guide to DNA and genome testing explains how test scope and reported findings fit together. If you later consult a database, retain the reference assembly and variant description so you can identify the correct record.
Compare methods with a short reading checklist
Start with the biological question, then work through the description in a consistent order:
- Identify the regions or markers directly examined and which types of variation the workflow addresses.
- Look for coverage, missing calls, filtering, and other quality information relevant to those targets.
- Separate directly measured results from imputed or otherwise inferred results.
- Check how detected differences become reported findings and what evidence supports the interpretation.
These questions are useful whether reading a paper or comparing the information supplied with DNA test kits. They avoid treating a marker count, a sequencing label, or a report length as a complete description of quality.
Genotyping is useful for determining alleles at chosen locations; sequencing can provide broader sequence information within its measured scope. The better fit follows from the question, target, and demonstrated performance. When the next step involves interpreting identifiers or annotations, our guide to genome database records shows how to preserve that context.
