Split Decision

With none of the public posturing that accompanied the initiation of the investigation, the Harris County District Attorney's Office has quietly dropped its probe into the forensic techniques employed by Dr. Elizabeth Johnson, the head of the medical examiner's DNA laboratory.

The D.A.'s Office subpoenaed Johnson to appear before a county grand jury in February and turn over her files on three rape cases and two murders ["Blood Feud," May 30]. To the consternation of prosecutors, the lab findings by the molecular biologist had resulted in the elimination of the prime suspects in each of those crimes. The D.A.'s initiative produced some high-tension courtroom drama as Johnson's attorney, Dick DeGuerin, squared off against District Attorney Johnny Holmes in an attempt to quash the subpoena. DeGuerin was unsuccessful, and Johnson was forced to produce the information.

But first assistant district attorney Don Stricklin notified DeGuerin last month that two independent reviews of Johnson's work had been completed and that "at this time" the D.A.'s Office does "not plan to pursue this matter any further." However, Stricklin did not completely let Johnson off the hook, telling DeGuerin there has been no published validation of the way Johnson performs DNA testing.

At issue was Johnson's technique in PCR, or polymerase chain reaction -- one of two types of DNA analyses. In most forensic DNA examinations, a process known as RFLP -- or restriction fragment length polymorphism -- is used. The PCR method is employed when there is a limited amount of DNA available for testing. It is not a new process, but Johnson claims a technique she helped perfect enhances the procedure so that it is possible to determine the gender of the donor of a DNA sample. She can do that, she says, by copying the sample of genetic material 35 times -- three more than the number of reproductions recommended by the manufacturer of the test.

Although Johnson's method of PCR analysis had passed muster with the California criminal courts system, some Harris County prosecutors had their doubts about its validity. So Holmes, in what he says was an effort to resolve the conflict, heeded the request of some of his assistants to have Johnson's work subjected to outside scrutiny.

But the findings of the two scientists who reviewed Johnson's files and tests -- Dr. Ranajit Chakraborty of the University of Texas Medical School and Hal Deadman, a retired DNA examiner with the FBI -- were inconclusive, according to Stricklin's letter to DeGuerin. Two of the five cases under review involved only the standard RFLP process, and Stricklin reported that the review "did not reveal any abnormalities in the RFLP cases." But the reexamination of the PCR process used in the other three cases "was not as definitive."

The three PCR analyses in question were performed on DNA samples collected in two murders and one rape. In those investigations, Johnson concluded that her testing either eliminated the prime suspect as the donor or that the DNA had to have been the victim's.

In DeGuerin's view, the D.A.'s Office concluded its probe in less-than-satisfactory fashion. "It's kind of a backhanded way of saying, 'There's egg on our face, and we were wrong,' " says the attorney, who contends that Johnson is owed an apology.

DeGuerin replied to Stricklin with a letter of his own, in which he contended that at least two available DNA studies -- including one published by the FBI in 1991 -- proved that reliable results can be obtained by using Johnson's enhanced PCR analysis. The lawyer suggested that the two reviewers employed by the D.A.'s Office should have been aware of those studies.

"Perhaps your experts were not as expert as you thought they were," DeGuerin wrote.

Moreover, says DeGuerin, Johnson's own report on her gender-based PCR protocol will be published in the January edition of The Journal of Forensic Sciences. (The editor of that publication has a policy of not commenting on upcoming articles.)

When contacted by the Press, Stricklin conceded the D.A.'s Office does plan to consult with its two reviewers about the studies cited by DeGuerin. However, he emphasized that there must be a good reason that the manufacturer of the standard PCR test suggests that DNA samples be amplified only 32 times, rather than Johnson's 35 times.

And although Stricklin confirmed that the D.A.'s review of Johnson's techniques has concluded for now, he declined to respond when asked if he has confidence in the work being done in the medical examiner's DNA lab. As for possible damage to Johnson's reputation, the prosecutor pointed out that it was DeGuerin -- not the district attorney -- who brought the investigation to the attention of the media.

Stricklin's letter to DeGuerin was dated July 22 -- five days before the arrival of Dr. Joye Carter, the county's new chief medical examiner. Carter replaces Dr. Joseph Jachimczyk, who retired last year just before a multitude of controversies surfaced at the ME's office.

Carter said she was not aware of the D.A.'s decision to drop the Johnson investigation, nor did she have an opinion about the work of the head of her DNA lab.

"We're reviewing all our methods here," says Carter, who recently hired a new assistant to oversee all the forensic labs at the county morgue. "And we're going to be meeting with all the [division] heads and go over what they've been doing.