DNA Microarray¶

A DNA microarray tests many known nucleic-acid targets in parallel through hybridization at indexed probe sites.

Version: v1 · 2026-10-31 · History
Domain-specific #: 14485
Domain group: Natural Sciences
Origin domain: Biology & Ecology
Subdomain: Genomics → Biology & Ecology
Aliases: Dna Chip

Core Idea¶

A DNA microarray is a spatially indexed nucleic-acid hybridization assay: known probes at mapped sites encounter labeled sample nucleic acid, complementary targets contribute site-specific signals, and an assay-specific analysis interprets the resulting signal pattern. Glass or another solid support is the physical implementation that preserves probe identity by location, not the whole abstract identity. Expression arrays and SNP genotyping arrays share this probe–hybridization–addressed-readout operation but infer different things from their signals. Neither is a general method for reading arbitrary unknown sequences.^[1]^[2]^[3]

How would you explain it like I'm…

DNA Matching Grid

Imagine a tray with lots of tiny spots, and each spot has its own sticky piece that only grabs one matching partner. Scientists pour in a glowing sample, and the matching bits stick to their spots and glow. Because they know which piece is on which spot, the glowing pattern tells them what was in the sample.

Matching DNA Spots

A DNA microarray is a tool with many tiny spots, each holding a known bit of DNA called a probe, arranged in a map so scientists know which probe is where. They label a sample's DNA or RNA so it glows, and wash it over the spots. Pieces that match a probe stick to it — this sticking is called hybridization — so matching spots light up. The pattern of light tells scientists things, such as which genes are active or which DNA letters a person has at certain places. It can only check for things the probes were designed for, not read any unknown DNA.

Addressed Hybridization Assay

A DNA microarray is a hybridization test organized by location: known probes sit at mapped positions, a labeled sample of nucleic acid (DNA or RNA) is applied, and target molecules complementary to a probe bind there and produce a signal at that spot. Analysis then interprets the pattern of signals. The glass slide or other surface is just how probe identity is kept tied to location; it's not the essence of the idea. Expression arrays measure how much of each gene's RNA is present, while SNP genotyping arrays detect which variant a sample carries — same basic operation, different conclusions. Neither can read an arbitrary unknown sequence; they only detect what their probes target.

A DNA microarray is a spatially indexed nucleic-acid hybridization assay. Probes of known sequence are immobilized at mapped sites; labeled nucleic acid from a sample is hybridized against them, and complementary targets generate site-specific signals. An assay-specific analysis then interprets the resulting signal pattern. The solid support — glass or another substrate — is the physical implementation that preserves probe identity by location, not the defining abstraction. Expression arrays and SNP genotyping arrays share the same probe–hybridization–addressed-readout operation but infer different quantities from their signals, such as transcript abundance versus allele identity. Because detection is limited to sequences complementary to the predesigned probes, a microarray is not a general method for reading arbitrary unknown sequences.

Structural Signature¶

Sig role-phrases: indexed probe array; labeled nucleic-acid sample; sequence complementarity; spatial signal; assay-specific inference.

Known probes are fixed at mapped positions.
A sample supplies DNA or RNA-derived target molecules.
Complementarity governs hybridization under assay conditions.
Signal by address reports binding at each feature.
Inference rules interpret signals as expression levels, genotype calls, or another designed target result.^[1]

What It Is Not¶

A microarray is not sequencing: it tests targets chosen in the probe design and may miss unrepresented variants. Fluorescence does not directly equal transcript abundance in every assay; expression estimates need calibration and comparison, while SNP genotype calling uses allele-specific patterns. A single nucleic-acid probe test lacks the array's many-address multiplex architecture.^[1]^[3]

Scope of Application¶

Schena and colleagues' cDNA microarrays monitored patterns of gene expression across many spots. The Wellcome Trust Case Control Consortium used a 500K mapping array for genome-wide SNP genotyping in an association study. The same probe–sample–signal chain appears in both, while their sample preparation and interpretation differ.^[2]^[3]

Clarity¶

The physical address matters because it identifies the probe: a bright spot is useful only if its immobilized sequence is known. Hybridization means complementary base-pair binding, not an automatic unique identification of any sequence that happens to emit light. Cross-hybridization and background can affect signal, so downstream calls require an assay-specific analysis.^[1]

Hybridization fluorescence needs controls and modeling before expression or genotype interpretation; a disease association requires a further case–control analysis. Diagnostic: what normalization, genotype-calling or association procedure connects an addressed signal to the claimed biological result?^[2]^[3]

Manages Complexity¶

The array converts thousands of separate target tests into one spatial readout. Its efficiency depends on designing an informative probe set and interpreting many noisy measurements. The design is both the source of multiplex power and a limit: an unrepresented transcript or variant cannot be inferred reliably merely because the chip is large.^[1]

Abstract Reasoning¶

Specify the biological question, choose probes for relevant known targets, prepare and label sample nucleic acid, hybridize and scan, then apply controls and a calling method appropriate to the assay. For expression, compare intensities under the study design; for genotype, classify allele-specific patterns. Check whether the target sequence exists on the chip before interpreting absence of signal.^[1]^[2]^[3]

Knowledge Transfer¶

Spatial indexing and complementary binding transfer between expression and SNP applications. The mapping from signal to result does not: one estimates transcript representation, the other calls genomic alleles. Clinical or causal conclusions require separate validation and cannot be read off a bright feature.

Examples¶

Constructed two-probe expression contrast¶

Consider a constructed interpretation example, not measurements from Schena's paper. Two known probes at sites A and B are complementary to two chosen transcript targets. After background correction and a declared common normalization, suppose the sample/reference fluorescence pairs are A: \(200/100\) and B: \(50/100\). The within-probe ratios are \(2.0\) and \(0.5\), or \(\log_2\) ratios \(+1\) and \(-1\). The spatial address says which designed target each ratio concerns; hybridization gives a relative signal contrast, not gene function, absolute molecule count or a causal mechanism. Schena and colleagues' actual original-study abstract reports simultaneous two-color differential-expression measurements for 45 Arabidopsis genes, but the accessible abstract does not provide these invented A/B values or a named gene-specific fold change.^[2]

Mapped back: known targets = transcript A and B; indexed probes = sites A and B; labeled RNA-derived sample/reference = two colors; observed toy signals = \(200/100\) and \(50/100\); bounded result = normalized relative ratios \(2.0\) and \(0.5\), not direct functional claims.

SNP mapping array¶

In the original WTCCC study, the Affymetrix GeneChip 500K Mapping Array Set was used for approximately 2,000 cases in each of seven diseases and roughly 3,000 shared controls. The case–control comparisons reported 24 independent association signals at \(P<5\times10^{-7}\), including nine for Crohn's disease. The array's hybridization features support genotype calls at designed markers; the 24 signals arise only after quality control and population-level statistical comparison, not from a bright probe alone. The accessible publisher page provides this abstract-level result, not a marker-by-marker raw fluorescence trace.^[3]

Mapped back: probe set = designed 500K marker features; sample = genomic DNA from cases/controls; site signal = evidence for allele-specific genotype calls; downstream comparison = 24 reported independent associations under the paper's threshold; inference boundary = association is not itself hybridization or causation.

Structural Tensions¶

Scale versus targeted coverage. Adding more indexed probes can measure more chosen targets in one assay, but increases design, manufacturing, signal-processing and multiple-comparison burdens; allocating finite features to one set of transcripts or variants can leave other sequences unrepresented. A large 500K array therefore broadens designed-marker coverage, not unknown-sequence discovery by itself. Diagnostic: does the probe set cover the target of interest at adequate specificity, and what cost or analysis burden accompanies broader coverage?^[1]^[3]

Structural–Framed Character¶

The identity is structural in its assay operation—known sequences, indexed addresses, complementary hybridization and location-bound signal—while its conclusions are framed by normalization, target design and the research question. Its evaluative weight differs between discovering a relative transcript contrast and making a reliable genotype call; neither follows from chip shape alone. Human laboratory practice selects probes, reference samples and controls, while institutional platforms such as the 500K mapping set standardize which variants are visible. The vocabulary travels literally between two-color expression and SNP studies because the same indexed hybridization operation persists even though their inference layers diverge. Importing “microarray” for any gridded sensor or data table is metaphor, not recognition of this nucleic-acid assay. Its character: a multiplex molecular measurement architecture with designed-target coverage and assay-dependent inferential limits.

Structural Core vs. Domain Accent¶

The skeletal relation is many parallel tests indexed by location. The domain-bound mechanism is nucleic-acid complementarity between known immobilized probes and labeled sample molecules, with signal-to-result analysis. The named entry fails the prime bar because an arbitrary spatial table lacks that molecular binding mechanism. Parallel addressed testing can also be used with non-nucleic-acid probes, which require their own detection mechanisms.

Hybridization supplies the molecular recognition operation, while sequencing is an alternative approach for certain questions.

Neighborhood in Abstraction Space¶

DNA Microarray sits in a moderately populated region (51^st percentile for distinctiveness): it has near-neighbors but no dense thicket of look-alikes.

Family — Molecular Biology & Genetic Engineering Methods (21 abstractions)

Nearest neighbors

Computed from structural-signature embeddings · 2026-10-31

Not to Be Confused With¶

Sequencing: reads sequence rather than only testing designed complementary probes. Expression array: a subtype using RNA-derived signal. SNP array: a subtype for specified variants. Association analysis: downstream statistics that can use genotype calls but is not the chip.^[1]^[3]

References¶

[1] National Human Genome Research Institute, Microarray Technology, glossary description of hybridization, readout and uses. registry ↩a ↩b ↩c ↩d ↩e ↩f ↩g ↩h

[2] Schena, Mark, Dari Shalon, Ronald W. Davis, and Patrick O. Brown (1995), “Quantitative Monitoring of Gene Expression Patterns with a Complementary DNA Microarray”, Science 270: 467–470, doi:10.1126/science.270.5235.467. Original abstract; Stanford-hosted original paper is image-only in this access pass. registry ↩a ↩b ↩c ↩d ↩e

[3] Wellcome Trust Case Control Consortium (2007), “Genome-wide association study of 14,000 cases of seven common diseases and 3,000 shared controls”, Nature 447: 661–678, doi:10.1038/nature05911. Original study abstract and reported analysis. registry ↩a ↩b ↩c ↩d ↩e ↩f ↩g ↩h