Why Science? | DNA Profiling, Genetic Evidence and Privacy

Section 1 of 36

1. DNA is biological information, not a verdict

DNA stores inherited instructions used by cells. A DNA profile examines selected variable regions to compare samples; it does not narrate a crime or prove when an activity occurred. Science makes the evidence powerful by making its limits visible. The profile, sampling process, statistics and case context must remain separate and connected.

Contents · Next section

Section 2 of 36

2. Chromosomes organise DNA

In human cells, long DNA molecules are packaged into chromosomes in the nucleus. Most body cells contain chromosome pairs, with one member of each pair inherited from each biological parent. A locus is a position in the genome. DNA profiling compares information at multiple selected loci rather than relying on one location.

Contents · Previous section · Next section

Section 3 of 36

3. Variation makes comparison possible

People share most of their DNA, yet some regions vary. Profiles often examine repeating sequences whose repeat counts differ among individuals. Variation does not make one person biologically better than another. In a profile, it provides distinguishable markers that can be measured and compared under a validated method.

Contents · Previous section · Next section

Section 4 of 36

4. A locus has possible alleles

An allele is a version found at a locus. A person may show one measured allele if both copies have the same repeat count, or two if they differ. Profile labels are measurements, not personality traits. A string of allele numbers should not be treated as a full genome sequence or a visible characteristic.

Contents · Previous section · Next section

Section 5 of 36

5. Repeats create measurable length differences

Short tandem repeats contain small DNA sequences repeated a variable number of times. After suitable laboratory processing, different repeat counts produce fragments of different lengths. Instruments can measure those fragments and assign allele labels. Each stage has thresholds and controls; the output is not a photograph of DNA itself.

Contents · Previous section · Next section

Section 6 of 36

6. A profile is deliberately limited

A standard forensic profile samples selected markers useful for identification. It is not normally designed to reveal a complete medical history, ancestry narrative or every gene. Even so, profiles are sensitive personal data. Limited scientific scope does not remove privacy responsibilities for collection, storage, access and later use.

Contents · Previous section · Next section

Section 7 of 36

7. Did You Know? A match is probabilistic

A matching DNA profile can be very informative, but the meaning is expressed through probability and competing propositions. The relevant question is not simply “Does it match?” It is how much more likely the observed evidence would be under one explanation than another, given the method, population information and sample conditions.

Contents · Previous section · Next section

Section 8 of 36

8. Collection starts the evidence chain

A swab, stain or tissue sample must be collected with appropriate permission, protective equipment and documentation. Location, date, collector and packaging matter. If the collection stage is uncertain, a precise laboratory measurement cannot repair the missing context. Chain of custody records who handled an item and when.

Contents · Previous section · Next section

Section 9 of 36

9. Extraction separates DNA from the sample

Laboratory extraction releases and purifies DNA from cells while reducing substances that can interfere with later steps. The amount and quality recovered can vary. A negative control helps reveal contamination introduced during processing. “DNA was extracted” should therefore be supported by controls and measurement, not assumed from the sample type.

Contents · Previous section · Next section

Section 10 of 36

10. PCR amplifies selected regions

The polymerase chain reaction can make many copies of targeted DNA regions. Primers define which regions are copied, and repeated temperature cycles drive the process. Amplification allows small amounts to be measured, but it can also amplify contaminating DNA. Controls, clean workflow and validated limits are essential.

Contents · Previous section · Next section

Section 11 of 36

11. Separation creates an electropherogram

Amplified fragments can be separated by size, often using capillary electrophoresis. Fluorescent labels allow an instrument to record peaks. Software helps assign allele labels according to calibrated size standards and analytical thresholds. A peak plot is processed measurement data, not an infallible fingerprint produced without judgement.

Contents · Previous section · Next section

Section 12 of 36

12. Allele calls need thresholds

Very small signals may be noise, while excessively high or imbalanced signals can also require attention. Laboratories validate thresholds and interpretation rules. Analysts examine controls and quality indicators. A classroom diagram usually removes this complexity; students should state that simplification before treating neat allele pairs as real casework.

Contents · Previous section · Next section

Section 13 of 36

13. Comparison starts locus by locus

An analyst compares the questioned profile with a reference profile at each usable locus. An unexplained difference can support exclusion, while agreement allows inclusion under the tested model. The comparison then needs a statistical weight. “Included” means not excluded by the profile; it does not mean proved responsible for an event.

Contents · Previous section · Next section

Section 14 of 36

14. Chain of custody preserves context

Each transfer, seal and storage condition should be documented. Chain of custody cannot prove that every laboratory result is correct, but gaps can weaken confidence that the tested item is the item originally collected. Good evidence is both analytically sound and traceably connected to its source.

Contents · Previous section · Next section

Section 15 of 36

15. Practise with invented profiles

The table uses fictional allele labels at four loci. Real systems use more markers, validated methods and statistical calculations.

Reference B differs at several loci. Reference A is not excluded, but this simplified comparison alone cannot identify a person or explain how DNA arrived.

Contents · Previous section · Next section

Section 16 of 36

16. Exclusion and inclusion are asymmetric

A reliable difference can exclude a reference as the source of a simple single-person profile. Agreement does not uniquely identify the source because other people can share alleles. The weight of inclusion grows when many informative loci agree and population frequencies are considered. State data quality before stating the conclusion.

Contents · Previous section · Next section

Section 17 of 36

17. Random match probability needs a model

Random match probability estimates how often a profile as discriminating as the observed one might occur in a relevant population under specified assumptions. It is not the probability that a named person is innocent. Confusing those questions is a serious reasoning error. Population data, relatedness and calculation method matter.

Contents · Previous section · Next section

Section 18 of 36

18. Likelihood ratios compare explanations

A likelihood ratio compares the probability of the evidence under two competing propositions. For example, one proposition may say a person contributed DNA; another may say an unknown unrelated person did. The ratio weighs the evidence between those propositions. It does not include every non-DNA fact or automatically become the probability of guilt.

Contents · Previous section · Next section

Section 19 of 36

19. Population databases support frequency estimates

Allele-frequency databases describe sampled populations and help calculate evidential weight. They must be suitable, quality-controlled and interpreted with population structure in mind. A database is not a list of suspects. Its purpose and governance should be distinguished from a database that stores identifiable personal profiles.

Contents · Previous section · Next section

Section 20 of 36

20. Mixtures are harder than single-source samples

A DNA mixture contains contributions from more than one person. Peaks can overlap, contributors may contribute unequal amounts and some alleles may drop out at low levels. NIST explains why mixture interpretation requires validated statistical methods. A neat classroom matching grid should never be presented as a substitute.

Contents · Previous section · Next section

Section 21 of 36

21. Transfer and persistence change meaning

DNA can be transferred directly or indirectly, and it may persist after the activity that deposited it. Finding DNA on an object supports presence under defined conditions, not necessarily when, why or by which route it arrived. Activity-level questions require context, experimental knowledge and competing explanations beyond the profile match.

Contents · Previous section · Next section

Section 22 of 36

22. Contamination is a testable risk

DNA from a collector, laboratory surface or another sample can be introduced unintentionally. Gloves, separated work areas, cleaning, controls and staff-elimination procedures reduce and detect risk. Claiming “contamination” without evidence is weak; claiming it is impossible is also weak. Examine controls, handling records and plausible routes.

Contents · Previous section · Next section

Section 23 of 36

23. Laboratory controls reveal failures

A positive control checks that the process can produce the expected result. A negative control should not produce a profile; an unexpected signal can flag contamination. Size standards and calibration support accurate allele calling. Controls do not guarantee perfection, but without them it is difficult to interpret a clean-looking result confidently.

Contents · Previous section · Next section

Section 24 of 36

24. Software and analysts both require validation

Software can model mixtures, estimate parameters and compare propositions, but it follows assumptions and settings. Laboratories validate systems using known samples and performance limits. Analysts need training and review. “The computer calculated it” is not a complete defence; the method, inputs, validation and reporting language remain inspectable.

Contents · Previous section · Next section

Section 25 of 36

25. DNA presence does not prove an activity

A profile can support that biological material from a person may be present. It does not by itself prove the person touched an object during a particular event, knew how it was used or committed an act. Separate source-level, activity-level and offence-level questions. Each step requires additional evidence and assumptions.

Contents · Previous section · Next section

Section 26 of 36

26. Relatives complicate independence assumptions

Biological relatives share more alleles on average than unrelated people. An identical twin presents an especially important limitation for many standard profiles. If a plausible alternative contributor is related, calculations and propositions must reflect that. Never infer family relationships from a classroom table or disclose sensitive suspicions.

Contents · Previous section · Next section

Section 27 of 36

27. Privacy continues after measurement

Genetic data can be persistent, identifying and relevant to relatives. Good governance addresses consent, purpose limitation, access, retention, security and deletion. A scientifically possible reuse is not automatically an ethically acceptable reuse. Privacy is part of research design, not paperwork added after collecting samples.

Contents · Previous section · Next section

Section 28 of 36

28. Do not collect classmates’ DNA

A school lesson can use paper profiles, simulations or synthetic data. It should not swab classmates, infer ancestry, test parentage or upload genetic information to consumer services. Consent among peers can be pressured and consequences can extend beyond the participant. Use invented identifiers and keep the inquiry about reasoning, not identity.

Contents · Previous section · Next section

Section 29 of 36

29. Map the full evidence chain

Draw collection → sealing → transport → extraction → quantification → amplification → separation → interpretation → statistical evaluation → report. Under each arrow, add a control or record. Then add a second chain for the case context. The map shows why a laboratory match is one important link rather than the entire argument.

Contents · Previous section · Next section

Section 30 of 36

30. Plot peaks without pretending they are people

An electropherogram plots signal intensity against fragment position or assigned allele. Label thresholds, controls and possible contributors. A bar chart of allele frequencies answers a different question. Keep units and denominators visible. Avoid decorative double-helix graphics when the task is to reason about measured peaks and probabilities.

Contents · Previous section · Next section

Section 31 of 36

31. Use Claim–Evidence–Reasoning cautiously

Claim: state inclusion, exclusion or limited support under named propositions. Evidence: cite loci, controls and statistical weight. Reasoning: explain why the pattern is more compatible with one proposition, then name transfer, mixture or sampling limitations. Do not leap from “not excluded” to “is the offender.”

Contents · Previous section · Next section

Section 32 of 36

32. What good Science tuition should build

Useful science tuition connects cells and inheritance to unfamiliar evidence. Students can interpret an invented profile, identify controls, distinguish match probability from guilt probability and critique a headline. Primary Science lays foundations in cells and fair testing; Secondary Science and O-Level Biology add chromosomes, DNA and experimental reasoning.

Contents · Previous section · Next section

Section 33 of 36

33. Choosing a school or programme

Check official subject offerings, laboratory teaching, research ethics and current programme details. A forensic workshop is an enrichment activity, not proof of a specialised pathway. Ask whether students learn data interpretation and privacy alongside technique. Verify admissions and qualifications directly because policies and programmes can change.

Contents · Previous section · Next section

Section 34 of 36

34. Careers require specialised responsibility

Forensic biology, molecular diagnostics, genomics, bioinformatics, laboratory quality, law and data protection intersect here. Roles have different education, accreditation and legal duties. School biology builds a foundation; it does not certify casework. Future professionals need validated methods, statistical literacy, ethics, documentation and careful communication.

Contents · Previous section · Next section

Section 35 of 36

35. Make the conclusion smaller and stronger

A responsible conclusion might say that a reference cannot be excluded as a contributor under a simplified single-source model. It should not say the reference “definitely left the sample during the event.” Include quality, statistical and contextual boundaries. Precision makes evidence more trustworthy, not less impressive.

Contents · Previous section · Next section

Section 36 of 36

36. The profile is powerful because it is bounded

DNA profiling joins molecular biology, measurement, probability and ethics. Its strength comes from traceable samples, validated methods, appropriate statistics and conclusions that stop where the evidence stops. That is why Science matters: students learn to welcome powerful evidence while protecting uncertainty, context and human privacy.

Try a final three-layer audit. At the analytical layer, ask whether the profile and controls are reliable. At the statistical layer, ask which propositions and population data were used. At the contextual layer, ask how, when and why biological material may have arrived. No layer can silently answer all the questions assigned to the others.

Parents and teachers can model careful language with fictional evidence. Replace “This proves who did it” with “This pattern excludes one reference and does not exclude another under the simplified model.” Then ask what additional non-DNA evidence would be needed. The smaller sentence is not evasive; it accurately marks what the exercise can support.

For revision, cover the labels on the evidence-chain map and rebuild them in order. Add one contamination control, one statistical warning and one privacy rule. Finally, explain why a match and an activity conclusion are different. If a student can do that without using real personal samples, they have learned both the power of the method and the responsibility that must accompany it.

End by checking every pronoun and probability statement. “It” should refer clearly to a sample, profile, proposition or person; those nouns are not interchangeable. “Rare profile” should not become “rare chance of innocence.” Precise language protects the reasoning from a famous statistical mistake and helps a reader see exactly which conclusion the evidence supports.

A final confidence check is to ask what observation would alter the interpretation: a failed negative control, an additional contributor, a related alternative source or a broken custody record. Evidence is strongest when the route to revision is explicit. That openness is exactly what distinguishes a scientific conclusion from a dramatic but untestable story.

Keep the human consequence in view as well. A laboratory label represents sensitive information about a person and sometimes their relatives. Accuracy, confidentiality and proportional use are therefore scientific-quality concerns as well as ethical ones. The best classroom success is not guessing an identity; it is explaining a defensible boundary with clarity and care.

Contents · Previous section · Continue to the Science Learning Hub