Chapter 8 — Key Takeaways: Advanced DNA

A one-page card. The chapter is the argument; this is the scaffold to hang it on.

The core claims

• DNA's clean single-source result is the easy case; most real casework is touch DNA, degradation, and mixtures. This is the chapter where the field's strongest method stops being automatic and starts requiring interpretation again — and interpretation is where bias and error re-enter.
• Sensitivity moved the battleground from "whose DNA" to "how did it get there." A perfect touch-DNA profile is a statement about a source of cells, never about an act. Secondary transfer (person → intermediary → object) can place an innocent person's DNA at a scene they were never near.
• Degraded and low-template results are graduated evidence — weaker than a clean profile in proportion to how little and how damaged the DNA was. "Partial match" means "match at fewer loci, with the rarity reduced accordingly." Never let it sound like "match."
• Mixtures are the hard problem of forensic DNA. The contributor count is a minimum, not a fact; a mixture rarely yields "this is his profile," at best "the evidence is X times more probable if he is a contributor" (the likelihood ratio — Chapter 9). "Inconclusive" is a valid, ethical result.
• mtDNA and Y-STRs work on hopeless samples but only narrow to a lineage, not a person — maternal for mtDNA, paternal for Y-STR. Strong for exclusion; never a one-in-a-billion individualization.
• Investigative genetic genealogy (IGG) is the chapter's triumph and its model of honesty: it generates an investigative lead from a distant relative's SNP profile, and gold-standard STR typing makes the actual identification. The new, less-validated step never does the courtroom work.

The method-validity verdict (where these sit on the NAS/PCAST spectrum)

What you can honestly say on the stand

"The DNA on this object is consistent with the defendant — that establishes whose cells are present. It does not establish how or when they were deposited; secondary transfer is possible, so the profile alone does not show the defendant handled the object."

For a mixture: "This is a mixture of at least two contributors; one component is consistent with [person]. I cannot treat the minor component as a single-source profile, and the number of contributors is a minimum, not a certainty."

For mtDNA / Y-STR: "This is consistent with the defendant and with his maternal [or paternal] relatives, and with roughly the fraction of unrelated people who share this type. It is a lineage match, not an individual identification."

For IGG: "Genetic genealogy produced an investigative lead. The identification offered as evidence is the conventional STR comparison between the crime-scene profile and the confirmed sample — that is the validated method doing the work."

Never say: "the DNA proves he was there," "the mtDNA identifies him," "genealogy identified the killer," or "low-template DNA is the same reliable test, just more sensitive."

Key terms (one line each)

• Touch DNA / trace DNA — DNA from shed skin cells on a handled object / any tiny biological deposit.
• Low-template DNA — too few cells; stochastic effects distort the profile.
• Secondary transfer — DNA arriving via an intermediary, not by direct contact.
• Allele dropout / drop-in — a true allele vanishes / a spurious allele appears, in low-template typing.
• DNA mixture — DNA from two or more contributors at once.
• Deconvolution — separating a mixture into its contributors (or assessing membership).
• mtDNA — high-copy, maternally inherited; survives degradation, shared along a maternal line.
• Y-STR — male-only Y-chromosome STRs; isolates a male contributor, shared along a paternal line.
• IGG — SNP profile → distant-relative match → family-tree → lead, confirmed by STR.
• SNP — a single-base genomic difference; the basis of genealogy matching.

Themes advanced in this chapter

• Theme 2 (validity spectrum): the spectrum runs inside DNA — clean STR, touch, mixture, lineage marker, and genealogy lead are different kinds of claim.
• Theme 3 (cognitive bias): interpretation (and thus bias) re-enters at the touch-DNA noise floor, in mixture deconvolution, and in genealogical tree-building; the fix is context management / blind interpretation (Ch. 31).

The cold case, after Chapter 8

The gas-can handle yielded a heat-degraded mixture: a major component consistent with the victim, plus an unknown minor contributor not in CODIS. A genealogy lead has been generated from the minor part. Nobody is included or excluded yet; interpreting the mixture (the likelihood ratio, and refuting the prosecutor's fallacy) is Chapter 9's job.

One-sentence summary

Push DNA to its limits — a few cells, cooked and partial, mixed with others, or belonging to a stranger CODIS never met — and the world's strongest forensic method becomes a graduated, interpretive one again, powerful but only as honest as the analyst who refuses to say more than "whose cells," "at least two," "a lineage," or "a lead."