eduKateSG · Why Science?
Define the exact gene–disease pair, inheritance model and evidence scope before scoring a relationship or trusting its classification
Choose the closest route, then return to the full index whenever you need the wider evidence chain.
Science learning becomes powerful when students can tell a measured pattern from a model, a statistical decision from a mechanism, and a useful lead from a finished conclusion. ClinGen’s Gene–Disease Clinical Validity curation process evaluates the strength of evidence supporting or refuting a claim that variation in a particular gene causes a particular monogenic disease. The current standard operating procedure compiles genetic, experimental and contradictory evidence from public sources and applies a structured framework plus expert judgement. The output is a classification for a specifically defined gene–disease relationship, including its mode of inheritance and disease scope. Gene–disease validity is not the same as classifying one variant, measuring dosage sensitivity, judging clinical actionability or diagnosing a person. This guide supports Primary Science, PSLE Science, Secondary Science, O-Level Science and STEM education as a longform evidence-reading route; it is not medical advice, a laboratory protocol, an admission promise or a career guarantee.
Related eduKate reading: ClinVar and evidence-aware variant classifications; Human Phenotype Ontology and structured clinical phenotypes; gnomAD and population allele-frequency context; Science Learning Hub; Education Hub; STEM Education; Careers by Subject Capability.
Primary and official sources checked on 11 October 2026: Official ClinGen Gene–Disease Validity activity; Current ClinGen Gene–Disease Validity SOP; ClinGen validity classification information; Original ClinGen evidence-based framework paper; 2026 Singapore–Cambridge O-Level Biology syllabus; MOE G2/G3 Lower Secondary Science teaching and learning syllabus; 2026 MOE G2 Computing syllabus. Package interfaces and recommendations can change, so reproduce an analysis only with the documentation for the installed release.
| Stage | Question | Evidence | Responsible output |
|---|---|---|---|
| Define | What exact gene, disease entity and inheritance model form the claim? | HGNC gene, disease identifier, scope and inheritance | A curatable gene–disease pair |
| Curate | What human genetic and experimental evidence supports it? | Case, segregation, case-control and functional evidence | A scored, traceable evidence set |
| Challenge | What contradicts the claim or changes its scope? | Population data, alternative causes, conflicts and lumping/splitting review | A balanced evidence record |
| Classify | What strength is justified after expert review? | Current SOP, classification criteria and approval | A dated validity classification, not a variant verdict |
Inside this guide
1–12 · Foundations and design
- 1. Start with a precise relationship
- 2. Separate validity from variant pathogenicity
- 3. Separate validity from dosage sensitivity
- 4. Separate validity from actionability
- 5. Define the disease entity
- 6. State the mode of inheritance
- 7. Use lumping and splitting deliberately
- 8. Gather public evidence systematically
- 9. Assess case-level evidence
- 10. Avoid duplicate cases
- 11. Evaluate de novo evidence
- 12. Evaluate segregation
13–24 · Models and estimation
- 13. Evaluate case-control evidence
- 14. Use population data as a challenge
- 15. Consider alternative explanations
- 16. Curate biochemical-function evidence
- 17. Curate expression evidence cautiously
- 18. Curate protein interactions carefully
- 19. Evaluate functional alteration
- 20. Evaluate model organisms
- 21. Evaluate rescue evidence
- 22. Score within the current SOP
- 23. Understand classification levels
- 24. Do not treat points as probability
25–36 · Diagnostics and interpretation
- 25. Recognise contradictory evidence
- 26. Distinguish Disputed from Refuted
- 27. Re-evaluate as evidence changes
- 28. Connect to Primary Science
- 29. Connect to Secondary Science
- 30. Connect to mathematics
- 31. Connect to computing
- 32. Connect to pathways
- 33. Did you know?
- 34. Write a reader-ready result
- 35. Build a validation ladder
- 36. Finish with a reproducible bundle
Section 1 of 36
1. Start with a precise relationship
ClinGen evaluates a gene paired with a defined disease and inheritance model, not a gene in isolation.
Working checkpoint — gene–disease pair: Write the full assertion before collecting evidence. Keep the independent sample, design, annotation and software version beside this decision so the evidence can be reconstructed.
For a classroom model, use invented counts and clearly label the activity as an analogy rather than a sequencing experiment.
Section 2 of 36
2. Separate validity from variant pathogenicity
Gene–disease validity asks whether variation in a gene causes a disease; variant classification asks whether one allele is disease-causing under a framework.
Working checkpoint — mode of inheritance: Keep these outputs in separate fields. Read the numerical result together with sample-level plots, effect size and a sensitivity check capable of weakening the headline.
For future study, connect the idea to biology, mathematics and computing while checking current official pathway information separately.
Section 3 of 36
3. Separate validity from dosage sensitivity
Dosage sensitivity evaluates whether loss or gain of copies is a disease mechanism for a gene or region. It is a distinct ClinGen activity.
Working checkpoint — genetic evidence: Cite the curation type used. Preserve the complete tested family and every excluded feature; a polished shortlist is not a substitute for provenance.
For scientific writing, prefer “supports”, “is associated with” and “is predicted to” when stronger causal verbs are not justified.
Section 4 of 36
4. Separate validity from actionability
Clinical actionability considers whether interventions are supported for a condition. A valid gene relationship does not automatically establish an action.
Working checkpoint — experimental evidence: Do not convert validity into care advice. Separate what the model estimates from the biological story proposed afterward, then name the extra evidence that story would require.
For students, the transferable habit is to ask: What was measured? Compared with what? What alternative explanation remains?
Section 5 of 36
5. Define the disease entity
Broad syndromes and narrow subtypes can produce different evidence sets. Current ClinGen procedures use disease nomenclature and expert scope decisions.
Working checkpoint — contradictory evidence: Record the disease identifier and included phenotypes. Treat identifiers, denominators and factor levels as scientific quantities: a silent mismatch can reverse an otherwise correct calculation.
For parents, a calm checkpoint is to ask how many independent samples support the claim and which assumption matters most.
Section 6 of 36
6. State the mode of inheritance
Autosomal dominant, autosomal recessive, X-linked and other models change which cases and variants support the claim.
Working checkpoint — validity classification: Keep inheritance with every evidence item. Keep the independent sample, design, annotation and software version beside this decision so the evidence can be reconstructed.
For a classroom model, use invented counts and clearly label the activity as an analogy rather than a sequencing experiment.
Section 7 of 36
7. Use lumping and splitting deliberately
Curators decide whether reported phenotypes represent one disease spectrum or separate entities. This can change classifications.
Working checkpoint — gene–disease pair: Document included and excluded phenotypes. Read the numerical result together with sample-level plots, effect size and a sensitivity check capable of weakening the headline.
For future study, connect the idea to biology, mathematics and computing while checking current official pathway information separately.
Section 8 of 36
8. Gather public evidence systematically
The SOP directs curators to assemble relevant genetic, experimental and contradictory evidence from available sources.
Working checkpoint — mode of inheritance: Save search strategies and review dates. Preserve the complete tested family and every excluded feature; a polished shortlist is not a substitute for provenance.
For scientific writing, prefer “supports”, “is associated with” and “is predicted to” when stronger causal verbs are not justified.
Section 9 of 36
9. Assess case-level evidence
Individual reports can support a relationship when variants are plausible and phenotypes, inheritance and alternatives are carefully evaluated.
Working checkpoint — genetic evidence: Trace each scored case to its publication. Separate what the model estimates from the biological story proposed afterward, then name the extra evidence that story would require.
For students, the transferable habit is to ask: What was measured? Compared with what? What alternative explanation remains?
Section 10 of 36
10. Avoid duplicate cases
The same person or family can appear in several papers, creating false replication.
Working checkpoint — experimental evidence: Cross-check pedigrees, variants and recruitment details. Treat identifiers, denominators and factor levels as scientific quantities: a silent mismatch can reverse an otherwise correct calculation.
For parents, a calm checkpoint is to ask how many independent samples support the claim and which assumption matters most.
Section 11 of 36
11. Evaluate de novo evidence
A confirmed de novo variant can be informative under a fitting disease model, but parentage, phenotype specificity and variant plausibility matter.
Working checkpoint — contradictory evidence: Record confirmation and scoring rationale. Keep the independent sample, design, annotation and software version beside this decision so the evidence can be reconstructed.
For a classroom model, use invented counts and clearly label the activity as an analogy rather than a sequencing experiment.
Section 12 of 36
12. Evaluate segregation
Co-segregation across informative relatives can support a relationship, while small pedigrees and ascertainment limit strength.
Working checkpoint — validity classification: Show the pedigree logic and calculation. Read the numerical result together with sample-level plots, effect size and a sensitivity check capable of weakening the headline.
For future study, connect the idea to biology, mathematics and computing while checking current official pathway information separately.
Section 13 of 36
13. Evaluate case-control evidence
Well-designed comparisons can support enrichment, but variant grouping, ancestry, technical matching and statistics define validity.
Working checkpoint — gene–disease pair: Inspect inclusion criteria and denominators. Preserve the complete tested family and every excluded feature; a polished shortlist is not a substitute for provenance.
For scientific writing, prefer “supports”, “is associated with” and “is predicted to” when stronger causal verbs are not justified.
Section 14 of 36
14. Use population data as a challenge
Variants too common for the proposed disorder and inheritance can weaken case evidence. Population references require release and ancestry context.
Working checkpoint — mode of inheritance: Match the exact allele and disease model. Separate what the model estimates from the biological story proposed afterward, then name the extra evidence that story would require.
For students, the transferable habit is to ask: What was measured? Compared with what? What alternative explanation remains?
Section 15 of 36
15. Consider alternative explanations
Another gene, environmental cause, mosaicism or phenocopy can explain a reported case.
Working checkpoint — genetic evidence: Record whether alternatives were reasonably excluded. Treat identifiers, denominators and factor levels as scientific quantities: a silent mismatch can reverse an otherwise correct calculation.
For parents, a calm checkpoint is to ask how many independent samples support the claim and which assumption matters most.
Section 16 of 36
16. Curate biochemical-function evidence
A gene product’s known function can support biological plausibility when relevant to disease mechanism.
Working checkpoint — experimental evidence: Distinguish direct evidence from pathway proximity. Keep the independent sample, design, annotation and software version beside this decision so the evidence can be reconstructed.
For a classroom model, use invented counts and clearly label the activity as an analogy rather than a sequencing experiment.
Section 17 of 36
17. Curate expression evidence cautiously
Expression in a relevant tissue is supportive context, but widespread expression rarely proves causation.
Working checkpoint — contradictory evidence: Report tissue, developmental stage and assay. Read the numerical result together with sample-level plots, effect size and a sensitivity check capable of weakening the headline.
For future study, connect the idea to biology, mathematics and computing while checking current official pathway information separately.
Section 18 of 36
18. Curate protein interactions carefully
Physical or genetic interaction with established disease proteins may support a pathway hypothesis.
Working checkpoint — validity classification: Require appropriate controls and biological relevance. Preserve the complete tested family and every excluded feature; a polished shortlist is not a substitute for provenance.
For scientific writing, prefer “supports”, “is associated with” and “is predicted to” when stronger causal verbs are not justified.
Section 19 of 36
19. Evaluate functional alteration
Patient-cell or non-patient-cell assays can show altered function, but model choice and rescue determine strength.
Working checkpoint — gene–disease pair: Separate observed alteration from disease causality. Separate what the model estimates from the biological story proposed afterward, then name the extra evidence that story would require.
For students, the transferable habit is to ask: What was measured? Compared with what? What alternative explanation remains?
Section 20 of 36
20. Evaluate model organisms
A model can recapitulate relevant phenotypes, yet species, allele and developmental differences limit transfer.
Working checkpoint — mode of inheritance: Match mechanism and phenotype before scoring. Treat identifiers, denominators and factor levels as scientific quantities: a silent mismatch can reverse an otherwise correct calculation.
For parents, a calm checkpoint is to ask how many independent samples support the claim and which assumption matters most.
Section 21 of 36
21. Evaluate rescue evidence
Restoring gene function or phenotype can strengthen a mechanism, provided design and specificity rule out simpler explanations.
Working checkpoint — genetic evidence: Inspect construct, controls and outcome. Keep the independent sample, design, annotation and software version beside this decision so the evidence can be reconstructed.
For a classroom model, use invented counts and clearly label the activity as an analogy rather than a sequencing experiment.
Section 22 of 36
22. Score within the current SOP
ClinGen uses structured points and caps to organise evidence, together with expert review and professional judgement.
Working checkpoint — experimental evidence: Use the current procedure rather than an old summary table. Read the numerical result together with sample-level plots, effect size and a sensitivity check capable of weakening the headline.
For future study, connect the idea to biology, mathematics and computing while checking current official pathway information separately.
Section 23 of 36
23. Understand classification levels
Current materials describe categories such as Definitive, Strong, Moderate, Limited, No Known Disease Relationship, Disputed and Refuted.
Working checkpoint — contradictory evidence: Quote the dated classification and its scope. Preserve the complete tested family and every excluded feature; a polished shortlist is not a substitute for provenance.
For scientific writing, prefer “supports”, “is associated with” and “is predicted to” when stronger causal verbs are not justified.
Section 24 of 36
24. Do not treat points as probability
A semiquantitative score supports consistent review; it is not a percentage chance that the claim is true.
Working checkpoint — validity classification: Explain the qualitative category and evidence. Separate what the model estimates from the biological story proposed afterward, then name the extra evidence that story would require.
For students, the transferable habit is to ask: What was measured? Compared with what? What alternative explanation remains?
Section 25 of 36
25. Recognise contradictory evidence
Population frequency, non-segregation, weak original reports or robust alternative explanations can oppose a relationship.
Working checkpoint — gene–disease pair: Curate contradiction with the same care as support. Treat identifiers, denominators and factor levels as scientific quantities: a silent mismatch can reverse an otherwise correct calculation.
For parents, a calm checkpoint is to ask how many independent samples support the claim and which assumption matters most.
Section 26 of 36
26. Distinguish Disputed from Refuted
Disputed indicates serious contradictory evidence; Refuted requires contradiction that significantly outweighs support under expert review.
Working checkpoint — mode of inheritance: Do not assign these labels casually. Keep the independent sample, design, annotation and software version beside this decision so the evidence can be reconstructed.
For a classroom model, use invented counts and clearly label the activity as an analogy rather than a sequencing experiment.
Section 27 of 36
27. Re-evaluate as evidence changes
New cases, functional data or contradictory observations can change a classification. Current records have approval and evaluation dates.
Working checkpoint — genetic evidence: Set a review trigger and preserve prior versions. Read the numerical result together with sample-level plots, effect size and a sensitivity check capable of weakening the headline.
For future study, connect the idea to biology, mathematics and computing while checking current official pathway information separately.
Section 28 of 36
28. Connect to Primary Science
A claim-evidence chart shows why one observation, several repeated observations and a controlled test support different strengths of conclusion.
Working checkpoint — experimental evidence: Use fictional traits and non-medical examples. Preserve the complete tested family and every excluded feature; a polished shortlist is not a substitute for provenance.
For scientific writing, prefer “supports”, “is associated with” and “is predicted to” when stronger causal verbs are not justified.
Section 29 of 36
29. Connect to Secondary Science
Inheritance, variation, gene function and experimental models converge in a gene–disease claim.
Working checkpoint — contradictory evidence: Separate human genetic evidence from experimental evidence. Separate what the model estimates from the biological story proposed afterward, then name the extra evidence that story would require.
For students, the transferable habit is to ask: What was measured? Compared with what? What alternative explanation remains?
Section 30 of 36
30. Connect to mathematics
Segregation, case-control statistics and evidence caps show how quantities enter a structured framework.
Working checkpoint — validity classification: Recalculate a simple published score without turning it into probability. Treat identifiers, denominators and factor levels as scientific quantities: a silent mismatch can reverse an otherwise correct calculation.
For parents, a calm checkpoint is to ask how many independent samples support the claim and which assumption matters most.
Section 31 of 36
31. Connect to computing
Identifiers, curation interfaces, versioned records and downloadable summaries make expert review reusable.
Working checkpoint — gene–disease pair: Validate gene, disease and inheritance fields. Keep the independent sample, design, annotation and software version beside this decision so the evidence can be reconstructed.
For a classroom model, use invented counts and clearly label the activity as an analogy rather than a sequencing experiment.
Section 32 of 36
32. Connect to pathways
Gene curation joins genetics, molecular biology, statistics, clinical expertise and data stewardship and can illuminate STEM education and career pathways.
Working checkpoint — mode of inheritance: Check current official training requirements separately. Read the numerical result together with sample-level plots, effect size and a sensitivity check capable of weakening the headline.
For future study, connect the idea to biology, mathematics and computing while checking current official pathway information separately.
Section 33 of 36
33. Did you know?
A Definitive gene–disease classification still does not mean every variant in that gene is pathogenic; the relationship and the allele answer different questions.
Working checkpoint — genetic evidence: Open the variant evidence separately. Preserve the complete tested family and every excluded feature; a polished shortlist is not a substitute for provenance.
For scientific writing, prefer “supports”, “is associated with” and “is predicted to” when stronger causal verbs are not justified.
Section 34 of 36
34. Write a reader-ready result
A strong summary names gene, disease, inheritance, evidence categories, points, contradictions, expert panel, classification and evaluation date.
Working checkpoint — experimental evidence: Make the classification traceable to evidence. Separate what the model estimates from the biological story proposed afterward, then name the extra evidence that story would require.
For students, the transferable habit is to ask: What was measured? Compared with what? What alternative explanation remains?
Section 35 of 36
35. Build a validation ladder
Verify scope and inheritance, deduplicate cases, evaluate genetic and experimental evidence, curate contradictions, apply the current SOP and obtain expert review.
Working checkpoint — contradictory evidence: Strengthen the least supported category. Treat identifiers, denominators and factor levels as scientific quantities: a silent mismatch can reverse an otherwise correct calculation.
For parents, a calm checkpoint is to ask how many independent samples support the claim and which assumption matters most.
Section 36 of 36
36. Finish with a reproducible bundle
Deliver the scoped assertion, identifiers, search log, evidence records, scores, contradiction table, SOP version, approval details and re-evaluation plan.
Working checkpoint — validity classification: Make every validity classification reconstructable. Keep the independent sample, design, annotation and software version beside this decision so the evidence can be reconstructed.
For a classroom model, use invented counts and clearly label the activity as an analogy rather than a sequencing experiment.
Contents · Previous section · Continue to the Science Learning Hub