Are DNA Tests as Accurate as We Thought?

Evidence

Posted Jul 23, 2008, 05:39 pm CDT
By Martha Neil

Trumpeted for years as virtually infallible, DNA tests may not be as accurate as we thought.

Questions raised by an Arizona lab worker who stumbled across several close DNA matches that seemingly defied the odds of a match, as described in court, have led to more extensive efforts to test DNA databases for accuracy in a number of states, reports the Los Angeles Times reported in detail this week in a lengthy article about the issue.

Those efforts are opposed by the Federal Bureau of Investigation laboratory that administers the national DNA testing system. It contends that the accuracy checks--which involve searches for matches between all DNA profiles, rather than matches to one specific profile--are misleading and unhelpful, and may violate the privacy of individuals whose profiles are being examined. It has also threatened to shut down labs that continue cooperating with such accuracy checks, the newspaper writes.

But analyst Kathryn Troyer points out that a number of labs have commonly matched suspects based on a nine-point ("location on chromosome" or "locus") comparison of their DNA profiles. In fact, a number of suspects may match at this level, leading to the possibility of a mistaken identification, she says.

At the request of Bicka Barlow, a San Francisco defense attorney defending a suspect in a rape-murder from 20 years earlier, a California court subpoenaed the Arizona database, seeking to determine how common such matches are among the population of suspects for which it has DNA samples.

"Among about 65,000 felons, there were 122 pairs that matched at nine of 13 loci. Twenty pairs matched at 10 loci. One matched at 11 and one at 12, though both later proved to belong to relatives," the Times reports.

The reason for the matches is clear when you understand the different way in which comparisons are made when searching for a suspect and searching for similar DNA profiles throughout the database:

"In a database search for a criminal case, a crime scene sample would have been compared to every profile in the database--about 65,000 comparisons," the Times article explains. "But Troyer compared all 65,000 profiles in Arizona's database to each other, resulting in about 2 billion comparisons. Each comparison made it more likely she would find a match."

Hat tip: How Appealing

Comments

Report Abuse

Posted by kay sieverding - 1 month, 2 weeks, 10 hours, 50 minutes ago

Right away, someone with $ should offer to pay for a full comparison of all the physical data that the matching persons are willing to provide. There is some funding and injunction power in the Crime Victims Act too.

The 10th Circuit had a case about a lab that produced the observations desired by the d.a. because that way they got more business and made more $..
Posted by associate - 1 month, 2 weeks, 9 hours, 8 minutes ago

Fraud is not the same thing as the scientific reliability of a DNA test.

I doubt anyone has ever been convicted on DNA alone. There’s always some additional items like credit card receipts, eye witnesses, and confessions that close the case. DNA is just a terrific starting point for those cases that seem random until you match DNA to one violent felon who happened to be in that town that day and happened to commit a similar crime previously. Things get a lot less random after that initial hit.

It’s always bad to rely solely on one piece of evidence, which is why that’s so rare.
Posted by kay sieverding - 1 month, 2 weeks, 9 hours ago

DNA evidence has probably been used to coerce a plea bargain that might involve an innocent person.
Posted by kay sieverding - 1 month, 2 weeks, 5 hours, 27 minutes ago

See “Gilchrist v. Citty, 173 Fed.Appx. 675 (10th Cir. 04/04/2006)”:

“Former Oklahoma City Police Department forensic chemist...performed forensic tests in OCPD’s investigations of violent crimes and testified in court regarding her analyses… commendations and letters of appreciation ...administration and operation of OCPD’s serology laboratory… new DNA laboratory ... early 1990s...outside consultant… evaluated...1999.... failed to properly document
....Mitchell v. Gibson, 262 F.3d 1036 (10th Cir. 2001)...characterized some portions of Gilchrist’s testimony as “without question, untrue” and concluded some of her courtroom statements were, “at least, misleading."… Laura Schile, an experienced DNA analyst.....problems came to light regarding evidence handling, packaging, and storage in the serology laboratory. ..evidence was missing...missing serology case files for the years ....exposed and unsealed evidence for a death-penalty case ....
Other evidence boxes had been wet, or were rotting, inappropriately sealed, or inappropriately marked....serology laboratory freezer malfunctioned and evidence was contaminated.....evidence was not stored and packaged properly. ...wasted tens of thousands of dollars on excessive, premature, or inappropriate equipment purchases for the DNA laboratory.*fn5 Importantly, Boshell’s report also alleged Gilchrist performed inaccurate forensic analyses, interpreted evidence incorrectly, and offered misleading testimony in criminal cases....
Gilchrist erred in her analysis of hair samples in the murder case....incorrectly determined the hair samples were consistent.....wrongly ruled out another suspect as a possible hair donor…
failed to disclose evidence and may have deliberately hidden test results...Association of Crime Scene Reconstruction had expelled Gilchrist for ethical violations....rape case....Gilchrist asserted the slides contained sperm and semen.....reexamined the slides...sperm was not present on any of the submitted slides....Special Agent Douglas Deedrick of the FBI ....Gilchrist’s laboratory notes frequently were incomplete or inadequate and asserted Gilchrist erred in identifying or interpreting samples in each of the five cases where she had used glass microscope slides to make hair or fiber comparisons....Gilchrist “made statements that went beyond the acceptable limits of forensic science.”
Posted by kay sieverding - 1 month, 1 week, 5 days, 7 hours, 20 minutes ago

Here’s that Justice for All act reference:

SEC. 206. ENSURING PRIVATE LABORATORY ASSISTANCE IN ELIMINATING DNA BACKLOG.
Section 2(d)(3) of the DNA Analysis Backlog Elimination Act of 2000 (42 U.S.C. 14135(d)(3)) is amended to read as follows:

(3) USE OF VOUCHERS OR CONTRACTS FOR CERTAIN PURPOSES-

(A) IN GENERAL- A grant for the purposes specified in paragraph (1), (2), or (5) of subsection (a) may be made in the form of a voucher or contract for laboratory services, even if the laboratory makes a reasonable profit for the services.

(B) REDEMPTION- A voucher or contract under subparagraph (A) may be redeemed at a laboratory operated on a nonprofit or for-profit basis, by a private entity that satisfies quality assurance standards and has been approved by the Attorney General.

(C) PAYMENTS- The Attorney General may use amounts authorized under subsection (j) to make payments to a laboratory described under subparagraph (B).’.

H.R.5107
Justice for All Act of 2004 (Enrolled as Agreed to or Passed by Both House and Senate)

Title 18 § 3771

Where are the Ernest Lawyer bloggers when we need them?

If “To Kill a Mockingbird” is such a favorite movie why don’t the people who liked the movie care about the subject of the wrongfully imprisoned?

Lawyers should have to tithe their time to do good law to keep their perspectives straight.
Or every day a good deed a mitzvah
Posted by DH Kaye - 1 month, 4 days, 20 minutes ago

This story starts with the statement that “DNA tests may not be as accurate as we thought.” It then contradicts itself, ending with the statement that “The reason for the matches is clear when you understand the different way in which comparisons are made ... .” Alas, it is not quite this simple.

For a more meaningful but slightly technical explanation of the issue, see Charles Brenner, Arizona DNA Database Matches, Jan. 8, 2007, http://dna-view.com/ArizonaMatch.htm. The law professors’ law-and-science blawg, at http://lawprofessors.typepad.com/science_law/2008/07/dna-database-wo.html, gives a less detailed explanation. (Shameless commerce disclaimer: I wrote the latter blog.)

Commenting has expired on this post.

ABA Journal - Law News Now

Evidence