“You submitted samples from an infant death case in Saint-Hyacinthe? When we talked, you mentioned that identity was an issue?”

Simone is a tiny woman with carrot hair, thick glasses, and the assertiveness of an earwig. Her excessive timidity results in the phrasing of most of her comments as questions. Drives me nuts.

“Yes.”

“We tried something a bit controversial. I hope that’s all right?”

“Controversial?”

“I thought you would want to know?”

“What did you try, Simone?”

“Are you familiar with BGA?”

“Ben Gurion Airport?”

“Biogeographical ancestry.”

“Tony Frudakis,” I said.

“Yes. And others. Though I believe Dr. Frudakis has abandoned this avenue of research?”

In the early 2000s, women were being murdered in Baton Rouge, Louisiana. Based on an FBI profile and one eyewitness statement, investigators sought a young white male as the serial killer. With no success. Frustrated, they turned to a molecular biologist named Tony Frudakis.

At that time DNA from a crime scene or victim could be used only for comparison to samples in CODIS, the Combined DNA Index System, a database containing approximately five million profiles. If investigators had a specimen but no suspect, they could run it through the database to see if it matched a sample on file.

While CODIS is good at linking unknown suspects to individuals already entered due to criminal activity, it’s useless for predicting ancestry or physical characteristics. And that’s no accident. When the National DNA Advisory Board selected the gene markers—the DNA sequences having known locations on chromosomes—for use in CODIS, they deliberately excluded those associated with physical characteristics or geographic origins. Couldn’t risk offending any ethnic group. No comment on that political reasoning.

DNAWitness, the test Frudakis developed and used in the Baton Rouge case, employed a set of markers selected precisely because they did disclose information about physical traits. Some were found primarily in people with Indo-European roots, others mainly in people of African, Native American, or South Asian heritage.

Frudakis told the Louisiana Multi-Agency Homicide Task Force that the perp they were seeking was 85 percent sub-Saharan African and 15 percent Native American. The Baton Rouge serial killer, linked by DNA to seven victims, turned out to be a thirty-four-year-old black man named Derrick Todd Lee.

“—distribution of these genetic markers has been associated with broad geographic regions, resulting in the recognition of BGA and the use of the markers as a major genetic component in the dissection of race. But you have to keep in mind that the diversity of these markers was complicated over the eons by historic events—population migrations, for example.”

While I’d been thinking about Frudakis, Annoux had gone into lecture mode. I heard no question marks when she talked science.

“People move around,” I said.

“Yes. Paleoanthropologists believe all modern humans are descended from populations that migrated out of Africa some two hundred thousand years ago. First they settled in the Fertile Crescent. Gradually, groups splintered off in every direction; eventually, some crossed the Bering Strait to America. With distance came reproductive isolation resulting in divergence of the gene pools.”