Fluorescence In Situ Hybridization¶

Preserve cells, tissues, or chromosomes in place, hybridize sequence-complementary probes to selected DNA or RNA targets, and read the retained fluorescent signal as target presence, number, arrangement, or spatial location.

Version: v1 · 2026-08-30 · History
Domain-specific #: 1856
Origin domain: molecular biology
Subdomain: molecular cytogenetics and spatial nucleic acid detection
Aliases: FISH, Fluorescent in Situ Hybridization

Core Idea¶

Fluorescence in situ hybridization (FISH) is an assay family that makes selected nucleic-acid sequences visible while retaining their position in a cell, tissue, chromosome preparation, or other spatially organized specimen. A probe with a sequence complementary to the intended DNA or RNA target is brought into conditions that permit selective hybridization. Unbound and weakly bound material is removed under controlled stringency. A fluorophore carried by the probe itself, or recruited through a labeled reporter, then makes the surviving probe–target hybrids observable by fluorescence microscopy or another validated fluorescence-reading system. The assay turns molecular complementarity into a spatially registered signal.^[1]^[2]

That combination is the identity. Hybridization supplies sequence selectivity; in situ preserves the target's relation to cellular, tissue, or chromosomal structure; fluorescence supplies the detectable reporter. Removing any one commitment changes the method family. Extracting nucleic acid before measurement loses the in-situ spatial frame. Replacing complementarity with antibody–antigen binding yields immunofluorescence. Replacing fluorescence with autoradiography or an enzyme-generated chromogen remains in situ hybridization, but not FISH. Merely imaging a fluorescent specimen has no sequence-specific hybridization commitment.

Historically, radioactive in situ hybridization established that complementary nucleic-acid probes could localize sequences in cytological material. Rudkin and Stollar demonstrated indirect fluorescent detection of in-situ DNA–RNA hybrids in 1977, and Bauman and colleagues demonstrated direct fluorescence localization with fluorochrome-labeled RNA in 1980.^[3]^[4] Langer-Safer, Levine, and Ward then used immunological detection to map cloned DNA sequences on Drosophila polytene chromosomes, while Pinkel and colleagues developed sensitive cytogenetic implementations for chromosome classification and aberration detection.^[5]^[6] These implementations differ, but all preserve the same probe–target–place–fluorescence operation.

FISH is therefore broader than one genetic test and narrower than in situ hybridization generally. It includes DNA-FISH and RNA-FISH, metaphase and interphase assays, locus-specific and chromosome-painting probes, direct and indirect reporters, single-target and multiplex designs, and research and clinical uses. It does not guarantee that a signal identifies a disease, a causal mutation, or a single nucleotide. The inference is limited by probe design, specimen integrity, hybridization specificity, optical and cytological resolution, controls, scoring rules, and the biological question.

Structural Signature¶

Locked operation: spatially preserved specimen + accessible DNA/RNA target + complementary probe + controlled hybridization and stringency washes + fluorescent reporter + spatially registered detection + validated interpretation -> target-specific presence, count, arrangement, or location claim.

The following roles are jointly diagnostic:

The preserved specimen and spatial frame — fixed cells, tissue sections, whole mounts, microbial communities, metaphase chromosomes, interphase nuclei, or another preparation in which biologically relevant position is retained.
The declared nucleic-acid target — a DNA locus, repeated region, chromosome-scale sequence collection, RNA transcript, microbial ribosomal RNA, or other sequence whose identity and intended inference are specified.
The complementary probe system — one probe or a designed set of DNA or RNA probes with sequence, length, labeling, and blocking choices intended to distinguish the target from homologous or repetitive background.
The accessibility treatment — fixation, permeabilization, denaturation, proteolysis, or other pretreatment balanced to admit the probe and expose the target without destroying the morphology needed for localization.
The hybridization environment — temperature, salt, pH, formamide or alternatives, probe concentration, duration, and competing nucleic acid adjusted so intended hybrids form preferentially.
The post-hybridization stringency step — washes and blocking that remove unbound or weakly specific material while retaining valid probe–target hybrids.
The fluorescent reporter branch — direct labeling, in which the probe carries a fluorophore, or indirect labeling, in which a hapten or other probe tag recruits a fluorescent antibody, avidin, secondary oligonucleotide, or amplification system.
The spatial readout — microscopy, image acquisition, spectral separation, registration to nuclear or tissue morphology, segmentation, and—where the claim requires it—spot counting, ratio calculation, distance measurement, or pattern scoring.
The control and calibration package — positive target controls, negative or non-target controls, internal control loci or cells, no-probe or detection-only controls when appropriate, and validated thresholds and exclusion rules.
The bounded claim — presence or absence, locus position, copy-number pattern, fusion or break-apart pattern, chromosome arrangement, organismal identity, or RNA abundance and location, stated only at the resolution and error rate the design supports.

The principal invariant is sequence-selective fluorescence that remains registered to an in-situ biological frame. A free fluorescent probe in solution is not a result; a bright spot unsupported by specificity controls is not a valid hybridization claim; a sequence call from extracted DNA is not in situ; and an image without a declared target and scoring rule is not an assay conclusion.

What It Is Not¶

Not in situ hybridization in general. ISH is the containing family. Radioactive, chromogenic, silver-enhanced, and other nonfluorescent detection systems preserve targets in place but do not instantiate FISH's fluorescent-reporting commitment.
Not generic nucleic-acid hybridization. Blots, arrays, and solution assays can use complementary probes. They need not retain the target's native cellular or chromosomal position, which is constitutive here.
Not fluorescence microscopy. Fluorescence microscopy is a detection platform. FISH adds a sequence-defined probe, a hybridization chemistry, stringency, and a target-specific interpretive model.
Not immunofluorescence. Immunofluorescence usually localizes antigens through antibody binding. Indirect FISH may use fluorescent antibodies to reveal a hapten-tagged probe, but the recognition event defining target specificity is still nucleic-acid complementarity.
Not sequencing or PCR. Sequencing reads nucleotide order and PCR amplifies a defined region, usually after extraction. FISH trades sequence-level coverage for direct spatial registration and cell-by-cell or chromosome-by-chromosome context.
Not a diagnosis by itself. A clinical FISH pattern becomes diagnostic only through a validated test, reference intervals or cutoffs, specimen and probe quality, appropriate nomenclature, and an interpretation joined to the clinical setting. The ACMG standard treats probe validation, scoring, controls, reporting, and equivocal results as explicit laboratory obligations.^[7]
Not chromosome painting, spectral karyotyping, break-apart testing, RNA-FISH, smFISH, flow-FISH, Q-FISH, CARD-FISH, MERFISH, or seqFISH as exact synonyms. These are probe, target, readout, amplification, or multiplex specializations inside or adjacent to the FISH family.
Not proof of molecular identity at arbitrary resolution. A visible spot can support only the probe-bounded and resolution-bounded claim. Cross-hybridization, overlapping signals, section truncation, autofluorescence, chromatic misregistration, and optical blur can support a false or overprecise interpretation.

Scope of Application¶

FISH is used across molecular cytogenetics, cell biology, developmental biology, pathology, microbiology, ecology, genomics, and spatial transcriptomics. In metaphase cytogenetics, locus-specific probes can place sequences on chromosomes, chromosome libraries can paint larger regions, and differently colored probes can expose rearrangements. In interphase nuclei, appropriately validated signal patterns can estimate chromosome or locus copy number without requiring a metaphase spread. In pathology, break-apart, fusion, enumeration, and amplification designs test declared structural or copy-number propositions in morphologically selected cells. In microbiology, oligonucleotide probes against taxon-informative ribosomal RNA can identify and spatially locate organisms in communities. In RNA-FISH, probe sets localize transcripts and may support cell-by-cell transcript counting or subcellular distribution analysis.^[8]^[9]

The shared identity does not make the protocols interchangeable. Metaphase chromosome preparations, formalin-fixed paraffin-embedded tissue, cultured cells, bacterial biofilms, embryos, and intact tissues differ in fixation, accessibility, autofluorescence, target abundance, morphology, optical depth, and controls. DNA-FISH often requires denaturing duplex DNA; RNA-FISH must protect RNA and may exploit multiple short probes against one transcript. A clinical enumeration assay needs validated cutoffs and scoring rules that a qualitative developmental-expression study may not. Multiplex spatial-transcriptomic systems add barcodes, repeated imaging, error-correcting codes, or sequential rounds far beyond classical single- or few-color FISH, while retaining the target-specific fluorescent in-situ readout.

The scope ends when spatial registration is immaterial or absent, fluorescent reporting is absent, complementarity is not the specificity mechanism, or the conclusion outruns the probe and readout. FISH can show that a probe-compatible sequence or transcript is present at a resolved location; it does not automatically show nucleotide-perfect identity, gene function, protein production, cell viability, causal pathogenicity, or genome-wide completeness.

Clarity¶

FISH compresses a complicated laboratory workflow into one intelligible question: where did a sequence-defined probe survive a controlled hybridization challenge, and what target claim does that pattern license? That question separates signal production from inference. The fluorophore says where reporter accumulated; probe complementarity and stringency say why it might correspond to the target; specimen morphology says where the target sits; controls say whether the assay worked and whether background could mimic it; the scoring model says what pattern counts as the claimed result.

A practical recognition test asks seven questions. Is a DNA or RNA target declared? Is a complementary probe or probe set declared? Is the specimen maintained in a spatial frame? Are accessibility, hybridization, and washes controlled? Is fluorescence physically linked to the hybridized probe directly or indirectly? Is signal interpreted relative to morphology and controls? Is the resulting claim no finer than probe, optical, and scoring resolution? If one answer is no, the work may be an adjacent method, an incomplete protocol, or an invalid inference rather than a valid FISH instance.

The direct/indirect distinction is particularly clarifying. In direct FISH, the fluorophore is covalently attached to the probe and the hybrid can be imaged after washing. In indirect FISH, the probe carries a nonfluorescent tag or initiator and a fluorescent binding reagent or amplification cascade reveals that tag. Indirect systems can increase sensitivity or flexibility but introduce extra binding, amplification, diffusion, and background mechanisms. Both branches remain FISH because the final reporter is fluorescent and its location is anchored to a complementary in-situ hybrid.^[10]

Manages Complexity¶

Without FISH, a claim about a nucleic-acid feature often splits into separate questions: Is the sequence present? Which cell contains it? Where in that cell or chromosome does it occur? How many target-pattern signals are visible? Do two regions remain together, split apart, or fuse? How heterogeneous is the specimen? FISH binds these questions to one spatially indexed readout. It preserves morphology and cell identity while using sequence complementarity to select the molecular feature, thereby avoiding the loss of heterogeneity that can occur when a bulk specimen is homogenized.

The abstraction also organizes method selection. A researcher first chooses the target proposition—location, copy-pattern, arrangement, taxonomic identity, or transcript distribution. That choice constrains probe architecture, specimen preparation, reporter branch, microscopy, controls, and scoring. A locus-position question may require metaphase spreads; an interphase enumeration test requires control probes and cutoffs; single-transcript counting needs a probe set, spot-separation assumptions, and a validated detection algorithm; tissue mapping requires morphology-preserving pretreatment and attention to depth and autofluorescence.

FISH does not eliminate complexity; it makes dependencies explicit. More aggressive permeabilization can improve probe access but damage morphology. Greater stringency can suppress cross-hybridization but also lose low-affinity true signal. Indirect amplification can reveal scarce targets but add nonlinear background. More multiplexing increases information per specimen while raising spectral, registration, decoding, and multiple-testing burdens. The role map makes those tradeoffs auditable instead of treating “the stain worked” as a sufficient criterion.

Abstract Reasoning¶

The structural signature supports several reliable inferences.

First, specificity is jointly determined, not possessed by the probe sequence alone. Sequence uniqueness, competitor DNA, hybridization temperature, ionic strength, formamide, wash stringency, target accessibility, and detection chemistry jointly shape which hybrids remain. A computationally unique probe can still yield misleading signal through specimen background or detection cross-reactivity; a probe overlapping repeated sequence can be rescued only by a justified blocking and interpretation design.

Second, absence of signal is underdetermined. It can mean absence or low abundance of target, but it can also arise from degraded nucleic acid, inaccessible target, failed denaturation, poor probe labeling, excessive stringency, photobleaching, focus or acquisition failure, or an invalid threshold. A positive control and an internal hybridization control separate biological absence from technical failure. Conversely, fluorescence is not sufficient evidence of intended target binding: autofluorescence, nonspecific probe retention, antibody background in indirect systems, bleed-through, dust, and segmentation artifacts can produce apparent positives. Negative, no-probe, nuclease, mismatched, or absent-target controls are selected according to the design.^[10]

Third, resolution limits the ontology of the conclusion. Optical resolution, section thickness, chromosome condensation, probe footprint, spot overlap, and image processing determine whether one can distinguish adjacent targets, individual transcripts, or rearrangement patterns. Two colors that overlap within the point-spread function demonstrate co-localization only at that effective resolution, not molecular contact. A chromosome-paint signal identifies material complementary to the library, not a nucleotide-resolved breakpoint. A split signal implies separation according to a validated distance rule, not an exact physical break coordinate.

Fourth, counts are models, not raw observations. Spot number depends on hybridization efficiency, allelic state, replication, nuclear truncation, ploidy, cell-cycle phase, segmentation, target clustering, and scoring rules. Clinical practice therefore establishes probe performance, normal controls, cutoffs, minimum cells, quality criteria, and reporting conventions rather than reading any atypical dot count as disease.^[7]

Knowledge Transfer¶

Knowledge transfers strongly within the FISH family at the level of roles and failure modes. A new specimen still needs preserved morphology, target accessibility, a probe whose specificity is defended, controlled hybridization and washing, a fluorescent reporter, an imaging plan, controls, and a bounded claim. Lessons about fixation–permeability tradeoffs, background suppression, spectral separation, positive and negative controls, and threshold validation can therefore guide movement among DNA-FISH, RNA-FISH, microbial FISH, whole-mount FISH, and tissue FISH.

Transfer is conditional, not recipe copying. A permeabilization that works in cultured cells may destroy an embryo or fail in a dense biofilm. DNA denaturation conditions can degrade RNA or morphology. A centromere enumeration control does not validate transcript detection. A clinical cutoff estimated in one specimen type or population does not transfer automatically to another. Direct probes and indirect amplification systems have different background and linearity. The correct transferable object is the dependency structure; concentrations, temperatures, incubation times, probe sets, microscopes, and thresholds require local validation.

The cross-domain portable residue belongs to existing primes rather than making FISH a prime. FISH instantiates Measurement by mapping a target attribute through a probe-and-instrument procedure to a controlled signal. It invokes Signal Detection Theory whenever thresholds trade false positives against false negatives, Measurement Uncertainty when counts or locations are reported, and Resolution Matching when probe and optical granularity must preserve the distinction the biological question needs. Its fluorophores, nucleic-acid complementarity, fixation chemistry, cytological morphology, and assay controls remain indispensable specialist cargo.

Examples¶

Locus mapping on a metaphase chromosome. A DNA probe complementary to a cloned sequence is hybridized to denatured metaphase chromosomes. A fluorescent signal is registered against chromosome morphology or band identity. The result supports a locus assignment at the probe and cytological resolution, not a base-pair coordinate. Pinkel and colleagues' high-sensitivity work showed that small chromosome-specific probes could yield visible, quantitative fluorescent signals in metaphase and interphase cells.^[6]

Interphase enumeration. A test probe and an internal control probe are hybridized in intact nuclei. The analyst counts signal patterns in morphologically and technically adequate cells under validated rules. The control helps distinguish locus loss or gain from hybridization failure and broader ploidy. Overlapping nuclei, truncated sections, split sister chromatids, weak signals, and borderline cutoffs require declared exclusion and reporting rules; a dot count alone is not self-interpreting.

Break-apart or fusion-pattern testing. Differently colored probes flank one locus or target two partner loci. Preserved adjacency, spatial separation, or fused signal patterns are scored against assay-specific distance and control rules. The pattern tests the designed rearrangement proposition; it does not discover all possible partners, exact breakpoints, or every genomic alteration.

Single-molecule RNA FISH. A set of singly labeled oligonucleotides binds along one RNA species. When enough independent probes occupy the transcript, their combined fluorescence forms a resolvable spot. Raj and colleagues showed that this architecture supports imaging and counting individual mRNA molecules; spot intensity, negative controls, probe-set behavior, and image analysis remain essential to distinguish transcripts from background.^[9] This is a powerful FISH specialization, not the definition of all RNA-FISH.

Microbial community localization. A fluorescent oligonucleotide probe complementary to a taxon-informative ribosomal RNA sequence is applied to fixed, permeabilized cells in a biofilm or environmental sample. The method can identify cells while retaining community architecture. Probe coverage, phylogenetic specificity, cell permeability, ribosome content, autofluorescence, and non-target controls bound the claim.

Nonexample: a fluorescent antibody stain. A fluorophore-tagged antibody localizes a protein in fixed cells. It is immunofluorescence, not FISH, because antigen recognition rather than nucleic-acid complementarity identifies the target.

Structural Tensions¶

Accessibility versus preservation. Fixation protects spatial organization and nucleic acid but can mask targets; permeabilization and proteolysis improve access but can destroy morphology, release RNA, or raise background.
Sensitivity versus specificity. More probe, permissive hybridization, lower stringency, or amplification may reveal scarce target while increasing off-target retention. Higher stringency may remove false signal and weak true signal together.
Signal amplification versus faithful quantity. Indirect reporters and enzymatic or hybridization-chain amplification increase detectability but can add nonlinear response, diffusion, extra binding backgrounds, or saturation. Presence/localization may remain valid when quantitative intensity does not.
Multiplex breadth versus separability. Additional colors or coding rounds increase targets per specimen but raise spectral overlap, bleaching, registration, decoding, and accumulated-error burdens.
Spatial context versus molecular breadth. FISH preserves cell and tissue context but usually interrogates preselected targets. Sequencing and arrays can survey more sequence space while commonly sacrificing direct native localization.
Resolution versus throughput. Higher numerical aperture, optical sectioning, denser sampling, and more fields can protect fine spatial distinctions but cost acquisition time, storage, and scoring labor.
Standardized scoring versus biological heterogeneity. Cutoffs permit reproducible clinical classification, yet sectioning, subclones, polyploidy, morphology, and borderline distributions resist reduction to one count. The scoring rule must preserve clinically or scientifically relevant heterogeneity without converting noise into a positive call.

Structural–Framed Character¶

FISH has a strong structural core but is domain-specific. The role sequence—preserve a frame, select a target by complementarity, remove weak matches, attach a visible reporter, and interpret the resulting spatial signal—can be expressed abstractly. However, literal application requires nucleic-acid strands, base-pair complementarity, fixation or specimen handling, hybridization thermodynamics, fluorophores, optical detection, cytological or histological morphology, and biologically grounded controls. Replacing those entities with analogous social or computational objects produces a metaphor, not another FISH assay.

The method is also partly framed by laboratory purpose. Research FISH may explore spatial expression or chromosome architecture; clinical FISH must add validated probe performance, quality systems, normal reference data, scoring thresholds, nomenclature, reporting, and handling of equivocal results. Those institutional requirements do not define the whole assay family, but they change when a fluorescent pattern may support action. The entry therefore keeps the molecular operation stable while stating use-specific validity requirements separately.

Structural Core vs. Domain Accent¶

Structural core: a target remains embedded in a meaningful spatial frame; a designed recognition element binds it selectively; non-target binding is challenged; a reporter converts surviving recognition into an observable signal; controls and a resolution model bound the inference.

Domain accent: the target is DNA or RNA; selectivity comes from Watson–Crick sequence complementarity under hybridization thermodynamics; spatial frames are cells, nuclei, chromosomes, tissues, organisms, or microbial communities; reporters are fluorophores detected optically; and validity depends on fixation, permeabilization, denaturation, stringency, probe design, spectral behavior, and biological or clinical scoring.

The domain accent is constitutive rather than decorative. It determines which mismatches destabilize a hybrid, how repetitive sequence is blocked, whether target survives preparation, whether probe penetrates, what fluorescence background exists, and what a signal pattern means. The portable recognition-and-readout form is already decomposable into more general catalog primitives. The irreducible residue—the exact integration of complementarity, in-situ preservation, fluorescence, and nucleic-acid interpretation—is why FISH merits a domain-specific node.

Measurement — FISH uses a probe, preparation, optical instrument, and scoring procedure to map a target attribute to a spatial signal and bounded result. This is the proposed single DAG parent because the assay cannot remain an assay after the measurement relation is removed.
Signal Detection Theory — signal thresholds, background distributions, and scoring rules trade sensitivity against false-positive rate, especially for low-abundance targets and automated spot detection.
Measurement Uncertainty and Observational Noise — hybridization efficiency, fluorescence variation, optical blur, sample truncation, and scorer variability limit counts and positions.
Resolution Matching — the probe footprint, optical point-spread function, specimen thickness, and scoring granularity must resolve the smallest distinction the intended claim protects.
Measurement and Disturbance — specimen preparation makes targets observable by changing the specimen; over-fixation, denaturation, digestion, and photobleaching can erase or distort what is measured.
Co-location — multicolor overlap can support co-location only at a declared effective resolution and does not by itself establish molecular interaction.

These relations explain FISH's general reasoning content. None covers the complete specialist identity, and no collection of them supplies nucleic-acid complementarity, in-situ specimen preparation, fluorescence-linked probe chemistry, and the method's established variants and controls.

Relationships to Other Abstractions¶

Current abstraction Fluorescence In Situ Hybridization Domain-specific

Parents (1) — more general patterns this builds on

Fluorescence In Situ Hybridization presupposes Measurement Prime

FISH uses a probe, preparation, optical instrument, and scoring procedure to map a target attribute to a spatial signal and bounded result.

Hierarchy path (1) — routes to 1 parentless root

Fluorescence In Situ Hybridization → Measurement

Neighborhood in Abstraction Space¶

Fluorescence In Situ Hybridization sits in a sparse region of the domain-specific corpus (89^th percentile for distinctiveness): few abstractions share its structure, so a faithful description tends to retrieve it precisely.

Family — Unclustered & Miscellaneous (1565 abstractions)

Nearest neighbors

Viability PCR — 0.83
Scanning Laser Ophthalmoscopy — 0.81
Structural Formula — 0.81
Comparative Genomic Hybridization — 0.79
Radiation Hybrid Mapping — 0.78

Computed from structural-signature embeddings · 2026-09-08

Not to Be Confused With¶

In situ hybridization (ISH): the containing method family; FISH is its fluorescence-detected branch. Chromogenic in situ hybridization: target hybrids are visualized through a colored precipitate or enzyme system rather than fluorescence. Immunofluorescence: fluorescent antibodies localize antigens; an antibody can participate in indirect FISH detection without replacing nucleic-acid hybridization as the specificity event. Comparative genomic hybridization: compares relative genomic representation; array CGH loses direct chromosome or tissue localization, while metaphase CGH uses a different competitive genome-wide design. PCR and sequencing: interrogate extracted or amplified sequence with different resolution and coverage. Radiation Hybrid Mapping: orders markers by co-retention after chromosome breakage and cell fusion; it does not directly localize complementary probes in cells or chromosomes.

Chromosome painting and spectral karyotyping are large-probe-set, chromosome-scale FISH variants. Break-apart and fusion FISH are pattern-specific clinical or research designs. DNA-FISH and RNA-FISH identify target classes. smFISH aims at single RNA molecules through multiple labeled oligonucleotides. M-FISH, SKY, MERFISH, seqFISH, Q-FISH, flow-FISH, fiber-FISH, and CARD-FISH alter multiplexing, coding, quantification, specimen geometry, amplification, or readout. They should be curated as recognized variants or specializations, not collapsed into exact aliases for every instance of FISH.

References¶

[1] National Human Genome Research Institute. “Fluorescence In Situ Hybridization Fact Sheet.” Updated 16 August 2020. registry ↩

[2] Levsky, J. M., and R. H. Singer. “Fluorescence In Situ Hybridization: Past, Present and Future.” Journal of Cell Science 116, no. 14 (2003): 2833–2838. registry ↩

[3] Rudkin, G. T., and B. D. Stollar. “High Resolution Detection of DNA–RNA Hybrids In Situ by Indirect Immunofluorescence.” Nature 265 (1977): 472–473. registry ↩

[4] Bauman, J. G. J., J. Wiegant, P. Borst, and P. van Duijn. “A New Method for Fluorescence Microscopical Localization of Specific DNA Sequences by In Situ Hybridization of Fluorochrome-Labelled RNA.” Experimental Cell Research 128, no. 2 (1980): 485–490. registry ↩

[5] Langer-Safer, P. R., M. Levine, and D. C. Ward. “Immunological Method for Mapping Genes on Drosophila Polytene Chromosomes.” Proceedings of the National Academy of Sciences 79, no. 14 (1982): 4381–4385. registry ↩

[6] Pinkel, D., T. Straume, and J. W. Gray. “Cytogenetic Analysis Using Quantitative, High-Sensitivity, Fluorescence Hybridization.” Proceedings of the National Academy of Sciences 83, no. 9 (1986): 2934–2938. registry ↩a ↩b

[7] Mascarello, J. T., B. Hirsch, H. M. Kearney, R. P. Ketterling, S. B. Olson, D. I. Quigley, K. W. Rao, J. H. Tepperberg, K. D. Tsuchiya, and A. E. Wiktor. “Section E9 of the American College of Medical Genetics Technical Standards and Guidelines: Fluorescence In Situ Hybridization.” Genetics in Medicine 13, no. 7 (2011): 667–675. Reaffirmed by ACMG in 2018; addendum published 2019. registry ↩a ↩b

[8] Pinkel, D., J. Landegent, C. Collins, J. Fuscoe, R. Segraves, J. Lucas, and J. Gray. “Fluorescence In Situ Hybridization with Human Chromosome-Specific Libraries: Detection of Trisomy 21 and Translocations of Chromosome 4.” Proceedings of the National Academy of Sciences 85, no. 23 (1988): 9138–9142. registry ↩

[9] Raj, A., P. van den Bogaard, S. A. Rifkin, A. van Oudenaarden, and S. Tyagi. “Imaging Individual mRNA Molecules Using Multiple Singly Labeled Probes.” Nature Methods 5, no. 10 (2008): 877–879. registry ↩a ↩b

[10] Young, A. P., D. J. Jackson, and E. L. Wyeth. “A Technical Review and Guide to RNA Fluorescence In Situ Hybridization.” PeerJ 8 (2020): e8806. registry ↩a ↩b