Anatomy of a Serial Killer

The work differs from the kind of DNA fingerprinting used to identify biological evidence left at a crime. It is extremely difficult—sometimes impossible—to extract conventional nuclear DNA markers from an old bone. The center has become skilled in extracting and analyzing a hardier but less-known source of DNA: that of the mitochondria that reside in our cells.Except for identical twins, each person’s nuclear DNA is unique. But each of us has another set of DNA located outside the cell’s nucleus and inside the mitochondria, the tiny organs that supply a cell with energy. We inherit mitochondrial DNA, known as mtDNA, directly from our mothers, and we share it with our siblings. It’s not unique, but mtDNA is enough to narrow the search for a victim’s family.Sansom spent almost an hour scrubbing and sanding the femur’s surface before attempting extraction. Few of the bones here contain marrow, which dissolves in the first two or three years after death. F2775.1EC had spent some 20 years in a box inside a police warehouse, so DNA would have to come from the scant cellular material inside the bone’s white scaffolding. She used a woodworker’s dremel to cut a rectangular window in the thickened area of bone just below the femur’s rounded head, where the thigh muscles once attached. Next she chilled, pulverized, and blended the sample inside a freezer mill loaded with sterilized ball bearings. Using an automated chemical process, she broke open the bone cells, released their genetic contents, and washed, concentrated, and purified the extract.

For genetic analysis, Sansom first had to increase the DNA to detectable amounts using a process called DNA amplification. Forensic software translated the results into a four-color graph of peaks and troughs. Drawing on her training and experience, she translated each graphic peak into one of the four nucleotide letters in the DNA alphabet. It took her about a week to process sample F2775.1EC.When the amplification signals aren’t clear, the chances for a reliable match plummet. In the worst case, the sequence data prove ambiguous, and workers must repeat the extraction and analysis. Sansom got her sequence on the first try. She uploaded it to the center’s DNA database. No hits. Then she uploaded the data to the FBI’s national missing-persons database. Again, no hits. Not yet.

Scaling the Backlog

In 2004 the center received a major investment to help realize Arthur Eisenberg’s goal of establishing a National Center for the Identification of Human Remains. It was the first of several National Institute of Justice grants given over a five-year period totaling more than $7 million. The center’s mission was to perform DNA testing on unidentified skeletal remains and “family reference” samples free of charge for any local or state law-enforcement agency that requested it. It’s now a clearinghouse at the heart of an effort to address the thousands of missing persons and unidentified remains discovered each year—what the justice department calls “America’s silent mass disaster.”

“The World Trade Center attack devastated this country with its massive loss of life,” Eisenberg says. “But if people only knew how many more unidentified murder victims there are . . . If you go back even 20 years, there are literally hundreds of thousands of families who have missing loved ones.” Even with generous funding, progress will ultimately hinge on making identifications cheaper, faster and more definitive, he adds. Laboratories such as the Center for Human Identification will be swamped now that more states mandate the collection of family-reference samples with missing-person reports. The center, Eisenberg says, must advance the technology used to identify human remains as it goes. By way of example, he cites a new program that can use broken bits of traditional nuclear DNA to identify weathered bones.

The tests scan some 40 lengths of highly fragmented DNA for single-nucleotide polymorphisms (or SNPs, pronounced “snips”), one-letter variations in the genetic code. The SNPs are then combined to create unique DNA fingerprints. If the center’s tests are successful—and Eisenberg says they’re making rapid progress—SNPs will allow forensic analysts to identify old bones more reliably than they can using mtDNA. “If SNPs pans out, it will be another revolution in how we deal with homicide,” the National Institute of Justice’s Morgan says. “There will no longer be a reason to have unidentified remains.”