Written by: Tara Luther, Promega with Deandra DaCosta, Palm Beach Sheriff's Office

A Seemingly Routine Investigation

In February 2023 a father, alarmed by a distressing phone call, used the Find My iPhone app to locate his 21-year-old daughter. When he arrived at the scene she reported that she had been sexually assaulted by her ex-boyfriend and law enforcement was contacted. Sikorsky’s lab received evidence from a Sexual Assault Kit (SAK) collected from the victim and a Physical Evidence Recovery Kit (PERK) collected from the suspect.

“When the samples first came into our lab, the case seemed like it would follow a typical workflow,” Sikorsky explained during her presentation. “We began by testing for male DNA using Y-screening, and everything seemed straightforward.”

However, what initially appeared to be a routine case soon took a complicated turn. The reference sample from the victim—a buccal swab typically used to generate a single-source DNA profile—returned an unexpected result. Instead of a straightforward profile, it came back as a robust two-person mixture.

The Confounding Results

“The moment we saw that mixture, we knew we had a challenge on our hands,” Sikorsky told the ISHI audience. A reference sample is meant to provide a single-source DNA profile from the victim, used as a known sample for comparison with the crime scene evidence. But in this case, the victim’s reference sample returned with two contributors, in a ratio of about 1:3, with the second contributor being highly prominent. Of note, only three loci in the sample showed more than two alleles.

“Our immediate thought was contamination,” she explained. “We paused all interpretation and requested a second reference sample from the victim.”

When the second sample also returned the same two-person mixture, the team ruled out contamination. “We knew something else had to be going on,” she said. With contamination no longer an option, the question became: why was the victim’s reference profile mixed?

A Critical Medical Revelation

Faced with this unusual result, Sikorsky and her team began considering other possibilities. “At that point, we started to think there might be a medical explanation. The analyst asked the detective to check into the victim’s medical history, and the answer came back quickly,” she shared at ISHI.

The victim had undergone a bone marrow transplant, with her biological sister as the donor. This information immediately changed the trajectory of the investigation. “That was the key to everything,” Sikorsky explained. “The victim was a genetic chimera.”

Julie Sikorsky presents at ISHI 35 San Antonio, Texas.

What Is Genetic Chimerism?

In Greek mythology, the Chimera was a creature made of several animals—a lion’s head, a goat’s body, and a serpent’s tail—symbolizing a being made up of disparate parts. In genetics, chimerism refers to an individual with two or more distinct DNA profiles within their body.

Chimerism most often occurs in humans following medical procedures like bone marrow transplants. When someone undergoes a bone marrow transplant their DNA profile may change over time. The recipient’s DNA profile can also vary between different cell types throughout the body, with some cells presenting as the recipient’s DNA type, the donor’s DNA type, or even a mixture of the two. In this case the victim’s buccal swab was not a single-source sample, but a mixture of her own DNA and her sister’s donor DNA.

The Forensic Challenge

For Sikorsky’s team, chimerism presented a new and complex challenge. “We had to rethink our entire approach,” she said. “The buccal swab was supposed to be the victim’s single source reference sample, but because she was a chimera, it turned into a two-person mixture.”

This complicated the statistical analysis. Forensic DNA software like STRmix™ requires a single-source reference profile to generate likelihood ratios (LRs). A mixed reference profile would confound the software, making it impossible to determine the statistical significance of the evidentiary profiles.

“Our first thought was to ignore the loci that had multiple alleles and treat the reference sample as single-source,” Sikorsky said. “But that wasn’t a viable solution. Ignoring data and not accounting for imbalance at other loci could lead to incomplete or inaccurate conclusions.”

Deductive Profiling: A New Approach

After consulting with experts and forensic forums, Sikorsky’s team decided to employ a more innovative approach. “We had to deduce a single-source profile for the victim,” she explained. “We treated the mixed reference sample as an unknown and used replicate analysis, conditioning the profiles on the donor’s DNA.”

By subtracting the donor’s contribution, the team was able to isolate what they believed to be the victim’s original DNA profile. “This gave us a profile we could work with,” Sikorsky said. “It wasn’t perfect, but it was scientifically rigorous enough to use for statistical analysis with the crime scene samples.”

Analyzing the Evidence

With the deduced victim profile in hand, the team proceeded to compare it with the crime scene DNA evidence. Their task was to determine whether the deduced profile, the donor profile, and/or the suspect’s profile could be excluded or was visually included in each sample.

“We were careful to scrutinize each piece of evidence,” Sikorsky recalled during her ISHI talk.

• Perianal/Anal Swab: The DNA analysis showed three contributors, likely from the suspect, the deduced victim profile, and the donor.
• Vaginal Swab: Similarly, the vaginal swab also revealed three contributors, with the deduced victim profile being more prominent than the donor’s profile.
• Neck Swab: In this sample there were two contributors to the DNA mixture, however, the deduced victim profile was excluded entirely, with the donor’s and the suspect’s DNA profile being the only detectable DNA.
• Penile Swab (from the suspect): When compared to the original mixed DNA standard from the victim, the victim could not be visually excluded from the DNA profile obtained from the suspect’s penile swabs. However, when the analyst later compared the deduced victim profile and the donor profile they were both excluded due to the possibility of drop out. This in turn presented challenges in generating a likelihood ratio for the original visual inclusion.

Overcoming Amplification Challenges

As Sikorsky explained during her ISHI presentation, her team faced additional challenges when amplifying the DNA. “We dealt with allelic dropout, where some alleles fail to amplify,” she said. “We attempted to re-amplify the penile swab at a higher target concentration to get more allele calls, but ironically this led to more dropout at certain loci.”

Despite these challenges, Sikorsky’s team was able to gather enough information to move forward. “It was a delicate balance between gaining more data and managing dropout, but in the end, we were confident in our results,” she said.

Transparent Reporting

Sikorsky emphasized the importance of transparency in the final report. “When you have a case this complex, it’s critical that every step is thoroughly documented,” she told the ISHI audience.

The report included details about the two-person mixture in the victim’s reference sample, the chimeric nature of her DNA, and the method used to deduce the victim’s pre-transplant DNA profile. ISO-compliant disclaimers were also added, explaining that the case interpretation relied on the medical history provided by the detective.

“The donor’s identity was protected throughout the report,” Sikorsky noted. “She wasn’t involved in the crime, so we referred to her only as the ‘donor profile.’”