The Recorder
By Evan Hill
January 15, 2008
Case Tests Cold Hit Link To '70s Murder
In the trial of a suspect in a 34-year-old San Francisco murder, the DNA match that brought the case back to life might also sow the doubt that frees the defendant.
In opening statements on Monday in the trial of John Puckett, attorneys on each side insisted that their interpretation of the DNA results would win the case. It appeared the trial would test how convincing a DNA match can be when it's based on evidence taken from a crime scene a generation ago.
Deputy Public Defender Kwixuan Maloof confidently dismissed the evidence as "some shabby DNA," saying one of the two samples collected from the crime scene completely excludes his client as a suspect, while the other is both unreliable and subjective.
Prosecutor David Merin painted Puckett as a repeat offender living in the city at the time of the murder and defended the forensic evidence briefly but vehemently, saying markers in the DNA evidence matched Puckett's at a number of points.
"John Puckett, for every location for which there is a marker, fits the crime-scene profile," Merin said.
Puckett, a three-time sex offender, was arrested in 2006 after San Francisco police matched his DNA to evidence in the 1972 rape and murder of Diana Sylvester, a 25-year-old nurse at UCSF hospital.
While Merin cast the DNA evidence as part of an overall picture that pointed to Puckett's guilt, an energetic Maloof focused on the genetic evidence as the reason his client should go free.
Maloof said the same assailant left semen in Sylvester's mouth and genitals and that even one of the prosecution's witnesses would testify that the genital sample completely excludes Puckett's profile.
He noted that while the FBI tests for 13 genetic markers common in all humans, the SFPD tested for only nine in the samples taken from Sylvester's mouth. Of those nine, Maloof said, only five and a half match Puckett's profile — not enough to overcome reasonable doubt.
"There are three possible ways to interpret the DNA from the mouth," Maloof said. "Two of the three possible interpretations exclude Puckett, and not one is scientifically better than another."
Merin told the jury that the chance of a random coincidence in the Puckett match, among American Caucasians, is one in 1.1 million.
When scientists make a full DNA match using all 13 markers, the probability of a random coincidence is usually about 100 trillion to one, said Henry "Hank" Greely, a Stanford law professor who specializes in law and biosciences. Greely said the statements by both attorneys indicate that the crime-scene DNA probably degraded between 1972 and the time it was compared with Puckett's. This may explain why police only tested against nine markets.
Greely said the 1.1 million odds given by Merin are "pretty low," meaning the chances of a coincidental match are higher than in most DNA comparisons.
Both Maloof and Merin are likely to spar throughout the trial over the quality and accuracy of the genetic match.
With Maloof sowing doubt in his opening statements, and Merin making no mention of a star witness who can place Puckett at the scene, the prosecution's case could rest on the ability to turn the ambiguity of DNA into a certainty.
The trial, before Superior Court Judge Jerome Benson, is expected to continue through the week.