You are using an outdated browser.
Please upgrade your browser
and improve your visit to our site.
Skip Navigation

How DNA Evidence Incriminated an Impossible Suspect

Not all DNA evidence is equally valuable or trustworthy

AFP / Stringer/ Getty

In 2009, a horrific murder occurred in a place rarely associated with violence: a Yale graduate scientific laboratory. On what was to be her wedding day, a graduate student’s body was found head down within a small mechanical chase behind a wall in the laboratory. As she fell, her underwear snagged and entangled on a vent pipe that spanned the length of the chase. Extensive DNA samples were taken from the victim, her clothing, and spaces around the chase. Testing revealed two profiles, one of which matched a co-worker later implicated in the crime through other evidence. But a second person’s DNA was also found, ominously recovered in significant quantities from samples that included the waistband of the victim’s underwear. When the profile was submitted to the DNA database, a match returned the name of a convicted offender living nearby.

Further investigation, however, turned up something mysterious. The database match suspect had died two years prior to the Yale attack. Stumped, investigators first ruled out an identical twin or other relative, as well as laboratory contamination errors. Ultimately, however, they learned that years earlier the offender had worked in construction. Specifically, he had spent one long, hot summer building the very mechanical chase in which the victim was found—and he had even made errors the first time around that required him to effectively rebuild it a second time. Even though the victim did not encounter that chase until years later, the fact that it was a space that was closed from ordinary traffic or regular cleaning, coupled with the building’s strict temperature and environmental regulation (as a result of its role as a scientific lab) helped preserve in pristine the large quantity of DNA the worker had left behind as he sweated in that space during the construction.

Amazingly, these cells rested undisturbed until the moment that they transferred to the victim as she fell through the cramped space. In other words, there was a DNA transfer—via skin and sweat cells, most likely—from the worker to the walls and pipes within the chase. And then, when the victim encountered those objects and that space years later, there was another transfer from those objects to her skin and underwear.

In the early days of DNA testing, examining samples was not unlike determining a person’s blood type. An analyst usually tested large drops of blood or semen stains, all visible to the naked eye. But forensic DNA testing can now yield results from samples with just a few cells, and routinely handles samples of several hundred cells. When you consider that over 10,000 cells can fit on the head of a pin, it becomes clear that the days of testing only large, visible stains are long past.

The ability to discern a DNA profile from a tiny amount of material is an indisputable benefit. No one wants a rapist to go undetected simply because he wore a condom, or a burglar to evade detection by wearing gloves. But with the ability to test the scantest of evidence comes a responsibility to recognize the limits on what inferences can be drawn from that evidence. It is one thing to arrive at a crime scene and test the smudges of blood on a windowpane broken by the robber; it is another to swab for any cell residue on a doorknob that was turned to get inside. 

The difference, of course, is that the bloody glass shard almost certainly came from the perpetrator, whereas any number of profiles might be obtained from a doorknob also touched by other people. That is because of a phenomenon about which little is known: DNA transfer.

Some people might consider DNA transfer a form of contamination. But that is not really a fair way to look at it. Contamination implies that a sample was somehow compromised along the way from collection to testing—that an uncontaminated sample would not have contained extraneous biological material. Contamination connotes lack of care on the part of a crime scene technician or laboratory analyst. With care and preventive actions, such as wearing protective gear or cleaning a work station, true contamination can be reduced or eliminated.

Transfer, on the other hand, is inevitable. It is not the product of accident or inadvertence, or sloppiness or malfeasance. It is simply life. Transfer is largely unavoidable unless we are all going to live in a bubble. It cannot be stopped through better training or education. And concern about transfer is the natural by-product of our ability to test samples so small as to be invisible to the human eye.

Researchers often distinguish between primary, or direct, transfer and secondary, or indirect, transfer. Primary transfer occurs when a person transfers his or her own DNA to another person or object by coming into contact with that person or object—for instance, when you kiss your spouse, your DNA is likely transferred via a small amount of skin or saliva cells. Similarly, when you pack your kid’s lunchbox, you leave your DNA all over it—from your handling of the items placed inside to the cells you deposit as you close the latch.

In the first decade or so of DNA testing, criminalists focused their attention on only the most obvious cases of primary transfer—on examining clear and unambiguous stains left directly by the perpetrator, such as the bloodied knife or semen-stained sheets—rather than looking for other, invisible cells left behind. It was not until 1997 that a study first suggested that DNA typing might be capable of recovering a profile from skin or other DNA-carrying cells on objects merely handled by the perpetrator.

Such trace, or touch, evidence, quickly garnered great attention. Since then, a series of studies has shown that DNA transfers quite readily based on even brief amounts of contact. Moreover, the amount deposited can range from just a few cells to a sizeable amount (recall that each cell contains about 6 pg of DNA, and most ordinary DNA typing methods today return reliable results with roughly 500 pg of template).

If primary transfer is the engine that turns the wheels of justice, by identifying perpetrators based on the DNA they leave behind, then secondary transfer is the cog that causes that wheel to grind to a halt. Secondary transfer occurs when DNA from one person is transferred to a person or object, even though the person never came into contact with that other person or object. Instead, the transfer occurs via an intermediary person or object. The problem is that what looks like primary transfer, and leads investigators to believe that the suspect was there, might in fact be secondary transfer.

Returning to the earlier example, suppose you give your spouse a kiss on the cheek. Later that day, your spouse meets an old friend, who plants a kiss on the same cheek. Secondary transfer may occur: your DNA may now be on the face or lips of the friend, even if you have never seen or met that person. The same scenario can happen with an object. For instance, when you send your child off to school with the lunchbox you packed, your DNA goes to school, too, and may end up on the teacher who collects the boxes to put them away, or on the shelf where they are stored, or on the table where your child eats—even though you do not come along personally or interact with any of those persons or surfaces.

The transference is called secondary because the original depositor of the DNA did not come into direct contact with place where the DNA was found. Using ordinals can also help additional suggest degrees of remove from the initial DNA donor. For instance, it constitutes tertiary transfer if your DNA starts on the lunchbox, goes from the box to your child’s teacher, and then is passed from the teacher to her aide. Tertiary helps signify that the DNA ended up in a place three steps removed from the initial contact with its owner. The ordinals used in the forensic context tend to start from the original source of the DNA sample, always defining its journey in reference to that originator.

But this language is also a bit confusing. DNA that transfers between persons or objects away from the originator is also, in a sense, directly transferring. Imagine that a source leaves a pool of blood on the floor. I step in it, which transfers that blood to the bottom of my shoe, which in turn rubs onto to a rung of a chair where I prop up my feet, which then transfers to the pants leg of a person who later sits in that chair. Each of those contact moments was direct, albeit at increasing remove from the original DNA depositor. Or if I speak, and tiny bits of saliva deposit my DNA on the ground in front of me—or I do my laundry and the mingling of the clothes distributes the shed cells contained therein—in both cases the transfer is direct. But, in the saliva example, the transfer is primary because the DNA emanated directly from the source, whereas in the laundry example it is secondary because it is moving among objects with which I initially came into contact.

Another important concept to grasp when considering transfer is the notion of persistence. Once DNA is deposited on a surface, be it a person or an object, how long will it stay there? How easily might it be erased or overwritten by contact from a subsequent person? Is the last person to come into contact with a person or item always the one to leave the dominant DNA profile in a sample? Such questions can be further complicated by asking whether the quality and duration of contact by each person influences the amount of transference to or from each party in the chain.

The Yale story illustrates how important it is, now that we have the capacity to link perpetrators to an offense by the presence of their DNA in criminal evidence, to understand completely how easy it is for such cells to appear in a sample with which the DNA donor has never come into contact. Absent a good alibi—in this case, the irrefutable proof of his prior death—the worker might have ended up implicated in the crime. His familiarity with the space, along with his prior record, might have been used against him to prove that he had special knowledge of a good place to dispose of the body. Unfortunately, we still know very little about DNA transfer, and one major entity that funds DNA studies—the National Institute of Justice—has not expressed much interest in learning more.

This article is an adapted excerpt from Inside The Cell: The Dark Side of Forensic DNA. Reprinted with permission from Nation Books.