In the 1990s, the dawn of DNA analysis in Orange County, forensic scientist Mary Hong was the queen of the cold case.

Arriving at the Orange County Crime Lab in 1985 — with her bachelor of science degree in criminalistics from Michigan State University — Hong became a forensic star in latex gloves.

In this Sept. 26, 2000, file photo, forensic scientist Mary Hong demonstrates the DNA testing procedures at the Orange County Forensic Science Center in Santa Ana.  (Photo by Eugene Garcia, Orange County Register/SCNG)

She led the way in solving stagnant, high-profile murder cases at the lab, the first local law enforcement DNA laboratory in the western United States. By 2005, she had testified more than 100 times in DNA cases.

The high point of her career: Hong linked four unsolved rape-murders in Orange County to killings in Northern California, helping to identify one of the worst serial killers in state history. Her work eventually aided in the capture of Golden State Killer Joseph DeAngelo, who would later plead guilty to 13 murders and dozens of rapes in the 1970s and ’80s.

In 1996, Hong submitted Orange County samples to the state’s fledgling DNA database and solved six murders right off the bat, according to the late author Michelle McNamara.

McNamara prominently featured Hong in the book “I’ll Be Gone in The Dark,” about the hunt for the Golden State Killer. Hong also has appeared in the television shows “Unsolved Mysteries,” “Cold Case Files” and “Dateline.”

‘Cold case closer’

Assigned in 1997 to a team formed to concentrate on old murders, Hong was, in the words of one top public defender, Orange County’s “cold case closer.”

She was a pioneer in the most advanced forms of DNA testing and analyzed the blood evidence that proved a dead woman found in an Arizona freezer was actually killed in Orange County. That finding meant that defendant John Famalaro should be tried in Santa Ana, not Arizona. Famalaro was given the death penalty in 1997 for the slaying of 23-year-old Denise Huber.

Crime writers described the lab-coated Hong as methodical and intrepid, having a “scientist’s dispassion.” In one case, a prosecutor speaking to the jury likened the work done by Hong and the crime lab to Galileo and Copernicus.

Former Orange County Crime Lab DNA specialist Mary Hong in 2012, who is accused by a defense attorney of altering the evidence to benefit prosecutors. (Photo by MINDY SCHAUER, ORANGE COUNTY REGISTER/SCNG)

‘Cooked the books’?

But now, Hong’s reputation as a scientist and leader among California forensic experts — she is the former president of the statewide association of criminalists — is under attack.

Hong, who left Orange County five years ago to run a lab for the California Department of Justice, faces allegations that she “cooked the books” in one recent murder case that unraveled midtrial. The case also has brought renewed attention to two other murder cases in which she was accused by the defense of tailoring her analyses to benefit the prosecution.

“She uses the science to come to a preconceived answer,” charged Assistant Public Defender Chuck Hasse. “She cooked the books.”

In response, Orange County District Attorney Todd Spitzer has launched a probe into local cases new and old in which Hong testified or provided forensic opinions. And the Los Angeles County District Attorney’s Office is considering a potential review of at least five cases in which she worked on behalf of the agency.

Hong did not respond to an email seeking comment.

Was DNA evidence tweaked?

At issue in the most recent Orange County case, involving a decades-old murder of a young Buena Park woman, is whether Hong deliberately tweaked DNA evidence in favor of the prosecution and then backed away from that analysis when it came time to testify in a second trial. Or was it just a matter of interpretation in a complicated field?

The answer lies in a forensic world that is not always black and white, but often gray.

“The data is the data, but the interpretation of the data can be skewed by the person interpreting it,” said Tiffany Roy, a former DNA lab worker, lawyer and forensics expert from West Palm Beach, Florida. “The lawyers think DNA is black and white and they don’t know how gray DNA can be.”

Scott Sanders, assistant Orange County public defender, has been sparring for more than 13 years with the local Crime Lab, which is operated by the Sheriff’s Department.

“For nearly two decades Hong was the Crime Lab’s ‘cold case closer’ and she clearly relished that role, which is the problem,” Sanders said. “The Crime Lab should not see itself as a loyal member of the prosecution team. A functioning crime lab must serve as a check on those who want to manipulate science for the sake of a conviction. When that’s not the case, ‘scientists’ become the most dangerous players in the criminal justice system.”

As president of the California Association of Criminalists, Hong warned in the group’s newsletter in 2009 that forensic scientists need to be vigilant in protecting the integrity of their work.

“We need to examine and analyze evidence without compromising its integrity,” she wrote. “We are required to have an in-depth understanding of the scientific processes that we utilize and be able to relay the significance of the results (and) …  we must ensure that the methods we use are validated and proper controls are in place to guarantee the results are accurate.”

Data reinterpreted in Buena Park case

But Hong’s integrity is under assault in the case against Daniel McDermott, accused of the rape and murder of an 18-year-old Buena Park woman in 1988. The case languished until 2009, when Hong conducted an analysis of DNA found on the victim’s right wrist.

After McDermott was arrested in 2012, Hong “reinterpreted” the data by removing genetic markers found in the first analysis. Those deleted markers would have shown McDermott was not a match or that the sample was, at least, inconclusive, said his defense attorney, Hasse.

Hong testified in 2016 that McDermott’s DNA was a match for the sample found on the victim’s arm. But the trial ended in a hung jury.

When it came time for a new trial this month, Hong told prosecutors she could not repeat her testimony on the DNA from the wrist. Without that testimony, the D.A. was forced to offer McDermott a plea bargain that allowed him to be immediately released with nine years time served. McDermott had been facing life in prison without parole.

‘Done deliberately’

Suzanna Ryan, a forensics expert hired by the McDermott defense, said Hong should never have reinterpreted her original analysis. Furthermore, Hong failed to get a second opinion on her reinterpretation, a violation of industry standards.

“I do believe it was done deliberately,” Ryan said.

She added, however, it is the only time she found evidence of cooking the data in the dozen or so analyses she reviewed by Hong.

“I don’t like what she did in this case, (but) I don’t think she is consistently changing data,” Ryan said.

Roy, the Florida expert, is not sure Hong did anything intentionally wrong in the McDermott case. “I don’t see anything that looks nefarious,” said Roy, who is not involved in the case.

At the request of the Southern California News Group, Roy reviewed a forensic report done by the defense as well as a recording of Hong explaining to prosecutors why she was having trouble with her earlier testimony.

Operating ‘on the fringe’

Roy said there was indeed a scientific basis for Hong’s deletion of the genetic markers — but barely.

Hong appeared to be operating on the fringe in disregarding the several markers that could have helped McDermott, labeling them as “noise.” Those markers could easily have been interpreted the other way, Roy said. She said Hong should have erred on the side of caution and included the markers, but was overly aggressive and may have succumbed to a bias for law enforcement.

Roy acknowledged forensic scientists often are pressured to overstep to give police and prosecutors what they want.

“Analysts are faced with this every day. The pressure is high to operate on the fringe,” Roy said. “These analysts are pressured to have the answers.”

In essence, scientists often want to be good teammates with law enforcement, she said. It’s called “confirmation bias.”

“They want to feel like they are helping the good guy,” Roy said. “You can see patterns that don’t exist, it’s in the human mind. It happens more than people realize.”

“These prosecutors and defense need to stay away from these scientists and let the science speak for itself.”

Sometimes when forensics experts on the witness stand go beyond what the analysis shows, no one in the courtroom knows enough to check them, Roy said. Defense attorneys don’t always retain their own forensics experts, especially when the client is poor or indigent.

“(Analysts) often know no one is checking on them,” Roy said. “There are a lot of overstatements. There’s a lot of pressure on these experts to fit into the prosecutors’ story.”

In the McDermott case, prosecutors didn’t talk to Hong for the second trial until the eve of her planned testimony. Hong said, in her recorded interview, she could not be sure the DNA standards had not changed since 2012, when she compared McDermott’s DNA to the wrist swab from the victim.

“I think I’m over-thinking this,” she said.

Hong told prosecutors she did not believe she had tailored her analysis to deliberately include McDermott. She said she didn’t purposely disregard the disputed markers as noise to incriminate McDermott. She just didn’t believe they were DNA, Hong told prosecutors.

She also said in their interview that she had tried earlier to set up a meeting with the prosecution team, but nothing was ever scheduled.

Other questionable cases

Even before the McDermott case collapsed, Hong was under fire for her contradictory testimony in two cases hinging on the age of semen found on the female murder victims. Both are Anaheim cold cases from the mid-1980s that were “solved” in the late 2000s.

In 1985 — long before an arrest was made — former Orange County criminalist Daniel Gammie concluded that semen found in murder victim Bridgett Lamon had not been deposited around the time of death, meaning she probably was not killed by the donor.

Former Orange County Crime Lab DNA specialist Mary Hong is reflected in a Tecan-Te-Shake, DNA extraction machine in 2012.  (Photo by MINDY SCHAUER, ORANGE COUNTY REGISTER/SCNG)

About 20 years later, the semen was tied to Lynn Dean Johnson, who was charged in the killing. Called to the witness stand, Gammie recanted his earlier analysis, saying the science had changed. He now believed the semen could have been deposited near the time of death, which would incriminate Johnson. Gammie’s then-supervisor, Hong, also testified to the same.

Despite the flip-flop, the Crime Lab never went back to review old convictions won using the antiquated science.

Months later, Hong wound up before the same judge to testify in another cold case with similar sperm evidence. This time she testified the semen could not have been deposited near the time of death, attorneys said.

Her analysis in that case allowed prosecutors to clear the victim’s boyfriend, whose semen was found in the victim’s underwear, and focus on suspect Wendell Lemond. Lemond was convicted and sentenced to 25 years to life.

Sanders, one of Johnson’s attorneys, said he used Hong and Gammie’s flip-flop to avoid the death penalty for his client. Lemond’s conviction is under appeal.

Attorneys in the two murder cases did not know about the contradictory testimony until years after the trials when they found out by happenstance.

Orange County lab officials have said the semen samples studied by Hong in the Lemond and Johnson cases were different enough to warrant her conflicting testimony.

In her 2009 newsletter message, Hong speaks about the hardships faced by forensic scientists and the need to “make the most of whatever evidence has been collected.”

Sanders responded that “making the most” isn’t always good.

“Mary Hong certainly lived by these words. Now the question is whether our criminal justice system cares.”