Transmission Disequilibrium Test¶
A family-based statistical test of whether an allele is transmitted from informative heterozygous parents to affected offspring more often than its untransmitted alternative under a specified null.
Core Idea¶
The transmission disequilibrium test (TDT) asks whether a marker allele is passed from heterozygous parents to affected children more often than the parents' alternative allele. The parent supplies an internal comparison: one allele was transmitted and the other was not. The observed imbalance is tested against a specified null. This is a family-based statistical test, not an allele-frequency comparison between unrelated patients and controls.[^ref-82c803cf48f3]
The original insulin-region study tested for linkage in the presence of association. A later oral-cleft study used affected parent–offspring trios for another phenotype and optionally modeled how exposures might modify transmission effects. That extension does not make an exposure variable a required part of every TDT.[ref-82c803cf48f3][ref-c8913410765e]
Scope of Application¶
The classical test needs affected offspring and marker genotypes that reveal an informative heterozygous parent and the transmitted allele. It can be applied to candidate markers for different affected traits. Its result must be read under the study's allele coding, family design and null, rather than as an automatic claim about the causal gene.[^ref-82c803cf48f3]
Spielman's insulin-dependent-diabetes example uses transmissions from families that include multiple affected offspring. Maestri and colleagues report 160 oral-cleft parent–offspring trios at markers associated with five candidate genes. Their original abstract supports the TDT application and optional effect-modifier analysis, but not marker-specific count tables or the complete regression model.[ref-82c803cf48f3][ref-c8913410765e]
Clarity¶
Separate a heterozygous parent's two possible marker alleles, the allele actually transmitted to an affected child, and the statistical inference from many such observations. A homozygous parent has no directional marker-allele choice for the classical comparison. The untransmitted allele is the matched alternative from the same parent; it is neither an unaffected sibling nor an unrelated control.[^ref-82c803cf48f3]
Excess transmission is evidence against a specified null under its assumptions. It does not alone prove that the marker causes disease. Spielman and colleagues checked unaffected-offspring transmissions to investigate segregation distortion as a different reason for inheritance imbalance.[^ref-82c803cf48f3]
Manages Complexity¶
The test condenses family genotypes into informative transmission comparisons. In the insulin-region analysis, 57 class-1/X heterozygous parents made 124 transmissions to diabetic offspring: 78 class 1 and 46 pooled class X, against 62 and 62 expected under the paper's no-linkage null. The 124 are transmissions, not 124 independent trios. The paper used its pooled biallelic statistic for this multiplex-family sample.[^ref-82c803cf48f3]
The summary only works when its conditions remain visible. Other allele codings and covariate models can require different calculations and interpretations; the original 78-versus-46 table cannot simply be reused for Maestri's interaction model.[ref-82c803cf48f3][ref-c8913410765e]
Abstract Reasoning¶
Identify the affected children, marker and parental genotypes. Select heterozygous parents, record which marker allele each passed to an affected child and which alternative remained, then compare the directional transmissions under the stated null or model. The original insulin-region paper reports a chi-square of 8.26 and p = .004 for its 78-versus-46 comparison.[^ref-82c803cf48f3]
Before interpreting such a result, ask whether the analysis is about overall excess transmission, disease linkage in the study's conditions, or an exposure-dependent effect. Also ask whether segregation distortion, ascertainment or genotyping problems could change the inference. A within-family control helps with population-composition artifacts of unrelated controls; it is not an assumption-free guarantee.[ref-82c803cf48f3][ref-c8913410765e]
Knowledge Transfer¶
The same family-transmission roles can be filled by different affected traits and markers, as in the insulin and oral-cleft studies. Their exact statistics and conclusions do not transfer automatically. A non-genetic matched-pair test may share a broader comparison idea without becoming TDT.[ref-82c803cf48f3][ref-c8913410765e]
TDT is a narrower kind of live Statistical Test: it has data, a null and an interpreted result, plus affected-offspring parental-transmission conditions. The broader test genus also includes tests that have no family genotype data.
Example¶
Insulin region and diabetes: Spielman's affected diabetic offspring and their genotyped parents form the family data. Fifty-seven class-1/X heterozygous parents supply informative choices. They transmitted class 1 in 78 observations and pooled class X in 46, compared with a 62/62 null expectation. The result is one disease-linkage application of the TDT roles.[^ref-82c803cf48f3]
Oral clefts: Maestri and colleagues applied TDT to 160 trios with cleft-affected children and their parents, examining markers near five candidate genes. Informative parental transmissions provide the within-family comparison. The study also extended the analysis with conditional-logistic effect modifiers and reported a maternal-smoking interaction near TGFA and TGFB3. The inspected abstract does not give the informative-parent counts or full model.[^ref-c8913410765e]
Relationships to Other Abstractions¶
Current abstraction Transmission Disequilibrium Test Domain-specific
Parents (1) — more general patterns this builds on
-
Transmission Disequilibrium Test is a kind of Statistical Test Domain-specific
TDT is a null-calibrated statistical test specialized to informative parental marker transmissions to affected offspring.
Hierarchy path (1) — routes to 1 parentless root
- Transmission Disequilibrium Test → Statistical Test
Neighborhood in Abstraction Space¶
Transmission Disequilibrium Test sits in a sparse region of the domain-specific corpus (96th percentile for distinctiveness): few abstractions share its structure, so a faithful description tends to retrieve it precisely.
Family — Unclustered & Miscellaneous (2721 abstractions)
Nearest neighbors
- Complex segregation analysis — 0.78
- Additive genetic effects — 0.77
- Infinitesimal model — 0.77
- Drosophila Hybrid Sterility — 0.77
- Scarr–Rowe effect — 0.76
Computed from structural-signature embeddings · 2026-10-17
Not to Be Confused With¶
- Unrelated case-control association: compares separately sampled groups rather than one heterozygous parent's transmitted and untransmitted alleles.[^ref-82c803cf48f3]
- Unselected-offspring transmission distortion: changes the affected-offspring sampling and the disease-linkage question; it can help check an alternative explanation.[^ref-82c803cf48f3]
- A causal-allele finding: a statistical transmission signal alone does not prove the tested allele is causal.[^ref-82c803cf48f3]
- An interaction extension: Maestri's covariate model adds an exposure-modification question; it is not the definition of the classical TDT.[^ref-c8913410765e]
References¶
[^ref-82c803cf48f3]: Spielman, Richard S., Ralph E. McGinnis, and Warren J. Ewens (1993), "Transmission Test for Linkage Disequilibrium, The Insulin Gene Region and Insulin-Dependent Diabetes Mellitus (IDDM)" (printed title uses a colon after “Disequilibrium”). American Journal of Human Genetics 52, 506–516. Original full article inspected, especially pp. 507–512, equation (5), and tables 5–6. https://pmc.ncbi.nlm.nih.gov/articles/PMC1682161/
[^ref-c8913410765e]: Maestri, N. E., et al. (1997), "Application of Transmission Disequilibrium Tests to Nonsyndromic Oral Clefts, Including Candidate Genes and Environmental Exposures in the Models" (printed title uses a colon after “Clefts”). American Journal of Medical Genetics 73(3), 337–344. Original indexed abstract and bibliographic metadata inspected; full article not inspected. https://pubmed.ncbi.nlm.nih.gov/9415696/