Chapter 7 — Self-Check Quiz

Twenty-five questions: multiple choice and short answer. Answer before opening the key. The answer key is in the collapsed block at the bottom.

Multiple choice

1. Forensic STR typing targets which part of the genome? - A. The protein-coding genes that determine traits like eye color - B. Short tandem repeats in non-coding regions that vary between people - C. The entire genome, sequenced base by base - D. Only the sex chromosomes

2. A locus is best described as: - A. The version of a sequence a person carries - B. A specific, named physical location in the genome that is targeted for typing - C. The instrument that reads DNA fragments - D. The national DNA database

3. An allele is: - A. A specific version (e.g., repeat number) found at a locus - B. A pair of chromosomes - C. A fluorescent dye used in PCR - D. A type of contamination

4. How many alleles does a person carry at a single autosomal STR locus? - A. One - B. Two — one inherited from each biological parent - C. Four - D. It varies from person to person

5. The polymerase chain reaction (PCR) is used in DNA typing to: - A. Sort DNA fragments by size - B. Copy targeted DNA regions exponentially until a trace becomes billions of copies - C. Determine a person's guilt - D. Store profiles in a database

6. In capillary electrophoresis, DNA fragments are separated by: - A. Color only - B. Size, because smaller fragments move through the polymer faster under an electric field - C. The analyst's visual judgment - D. Their guilt-relevance

7. An electropherogram is: - A. The chemical that breaks open cells - B. The visual readout of a profile, plotting fluorescence against fragment size, with alleles as peaks - C. A national database - D. A type of PCR primer

8. Two profiles differ at one locus (evidence (15,16); suspect (14,17)) but match at all others. This is, barring rare exceptions: - A. A match - B. A strong inclusion - C. An exclusion — the suspect is not the source - D. Inconclusive and meaningless

9. A CODIS "hit" is best understood as: - A. Conclusive proof of guilt - B. An investigative lead that must be confirmed by direct re-typing of a fresh reference sample - C. A name and photograph appearing on screen - D. A random match probability

10. The random match probability (RMP) answers which question? - A. Given a matching profile, how likely is the defendant innocent? - B. Given that a random unrelated person is not the source, how likely is it they would coincidentally match the evidence profile? - C. How often does the lab make an error? - D. How long ago was the DNA deposited?

11. The "prosecutor's fallacy" is the error of: - A. Stating the RMP as a fraction - B. Equating "the chance of this profile in a random innocent person" with "the chance the defendant is innocent" - C. Running a negative control - D. Typing too many loci

12. Per-locus match probabilities can be multiplied together to get the profile frequency because: - A. The loci are inherited largely independently of one another - B. Multiplication always makes numbers smaller - C. The database requires it - D. The defendant agreed to it

13. A one-in-a-billion RMP does not account for: - A. The rarity of the profile in the population - B. The number of loci typed - C. Laboratory error and close relatives - D. The fluorescent labels

14. Why is DNA placed at the top of the validity spectrum (NAS 2009 / PCAST 2016)? - A. Because juries trust it - B. Because it has an understood biological basis, a quantified error structure, and an objective core measurement - C. Because it is the oldest forensic method - D. Because it never makes mistakes

15. The transfer problem refers to: - A. Moving samples between labs - B. DNA being deposited at a location where its source was never present (e.g., secondary transfer) - C. Transferring a case to another jurisdiction - D. Converting an electropherogram to a profile

16. A pristine, single-source DNA match establishes, by itself: - A. That the source person is guilty - B. When the DNA was deposited - C. Only that the cells came from that person - D. How the DNA got to the object

17. Historically, the technique Alec Jeffreys first developed (1984) that was later largely replaced because it needed more, intact DNA, was: - A. Capillary electrophoresis - B. RFLP ("DNA fingerprinting") - C. CODIS - D. Touch DNA

18. The first criminal use of DNA evidence (the Pitchfork case) is emblematic because the test first: - A. Convicted the obvious suspect - B. Excluded an innocent suspect before it identified the guilty man - C. Failed entirely - D. Identified the killer from a database

19. Which is an honest expert statement? - A. "There is a one-in-a-billion chance the defendant is innocent." - B. "This profile is found in approximately one in a billion unrelated individuals in the relevant population." - C. "The database identified the defendant as the perpetrator." - D. "The DNA proves he did it."

20. A negative (reagent-blank) control is run primarily to detect: - A. Close relatives - B. Contamination introduced during processing - C. The prosecutor's fallacy - D. Peak height

Short answer

21. In one sentence each, explain why a single clean mismatch can exclude a suspect with near-certainty while a 20-locus match only supports that the suspect is the source.

22. Explain why "every gain in sensitivity comes with a cost" in forensic DNA — what is the cost, and why does it grow as sensitivity increases?

23. A doorknob profile matches a suspect at a one-in-a-trillion RMP. List the three questions §7.6 says the match cannot answer, and state in one sentence why this matters to a verdict.

24. Distinguish the coincidental-match risk from the laboratory-error risk. Which is realistically larger, and why does typing more loci not reduce the second?

25. Why does the chapter call DNA both the forensic yardstick and the instrument that proved the other methods wrong? Connect your answer to the post-conviction exonerations of Chapter 6.

Answer key (click to expand) **1.** B — Typing targets variable short tandem repeats in non-coding DNA, not coding genes. **2.** B — A locus is a named physical address in the genome; the version there is the allele. **3.** A — An allele is the specific version (e.g., repeat number) at a locus. **4.** B — Two, one from each parent; if both are identical the locus shows a single value (homozygous). **5.** B — PCR copies targeted regions exponentially. **6.** B — Separation is by size under an electric field; smaller fragments migrate faster. **7.** B — The electropherogram plots fluorescence vs. fragment size, alleles appearing as peaks. **8.** C — A clean, reproducible mismatch at even one locus is an exclusion. **9.** B — A CODIS hit is an investigative lead; it stores numerical profiles, not identities, and must be confirmed by direct re-typing. **10.** B — The RMP is conditioned on the person being a random unrelated non-source; it is *not* the probability of innocence. **11.** B — The prosecutor's fallacy transposes the conditional, equating the RMP with the probability of innocence. **12.** A — Independence of the loci is what licenses multiplying per-locus frequencies (with statistical corrections). **13.** C — The RMP describes coincidental matching only; lab error and close relatives are separate, and unaffected by adding loci. **14.** B — Understood biology, quantified error structure, and objective core measurement put DNA at the top of the spectrum. **15.** B — Transfer (especially secondary transfer) puts a person's DNA where they were never present. **16.** C — A match establishes the *source* of the cells, nothing about timing, mechanism, or guilt. **17.** B — RFLP, which needed more and more-intact DNA and was largely replaced by PCR-based STR typing. **18.** B — In the Pitchfork case DNA first excluded an innocent suspect, then helped identify the guilty man. **19.** B — Only B states the rarity of the profile without committing the prosecutor's fallacy or overstating to "guilt." **20.** B — The negative control catches contamination carried through the process. **21.** A mismatch means the suspect lacks an allele the true source must have, so (barring lab error) the suspect cannot be the source; a full match means the suspect *could* be the source, but others with the same profile could be too, so the match must be weighed by the RMP and against transfer, relatives, and error. **22.** The cost is that greater sensitivity also detects irrelevant DNA — stray cells, contamination, secondary transfer; as the method reads ever smaller amounts (down to a few dozen cells), the chance that the detected profile has nothing to do with the crime rises. **23.** It cannot say *when* the cells were deposited, *how* they got there, or *what the person was doing*; this matters because a strong source attribution is not a statement about guilt, and innocent transfer or timing can fully explain a match. **24.** The coincidental-match risk is the chance an unrelated stranger happens to share the profile (the RMP, often astronomically small); the lab-error risk is the chance of a swap, contamination, or mislabel (a real, human rate). The lab-error risk is realistically larger and does not shrink with more loci, because it is independent of profile rarity. **25.** DNA is the *yardstick* because its understood, quantified validity is the standard against which weaker pattern methods are judged and found wanting; it *proved the other methods wrong* because every post-conviction DNA exoneration (Ch. 6) overturned a conviction built on a less reliable method (hair, bite marks, eyewitness, confession), demonstrating empirically how often those methods erred.