Weekend Diversion: Have a seat, CSI

New technology has led to breakthroughs in practically every aspect of our lives since the dawn of the industrial revolution. In nearly every case, it's for the betterment, in some way, of society. (And yet, I like my music best when it's acoustic, unprocessed, and barely produced at all.) Take a listen to Song for Roy by the amazing Sam Bush (with help from Emmylou Harris) about his late bass player, Roy Huskey, Jr.

One of the more memorable technological advances in my lifetime came in the mid-1990s, when DNA testing/evidence became a very controversial part of our culture with a certain news event.

In the fifteen-or-so years since then, DNA evidence has gone from a very expensive, laborious, and suspect new technology to the gold standard for crime-scene evidence. Convictions have been both made and overturned solely based on new DNA evidence. After all, DNA is your genetic fingerprint, so unless you've got an identical twin, there's nobody out there with the same DNA as you.

But a recent study showed that -- just as the rise in computers brought along hackers -- DNA evidence can now be completely fabricated! From this recently published article, here's a bit from their abstract, with my emphasis in bold.

DNA evidence is key to the conviction or exoneration of suspects of various types of crime, from theft to rape and murder. However, the disturbing possibility that DNA evidence can be faked has been overlooked. It turns out that standard molecular biology techniques such as PCR, molecular cloning, and recently developed whole genome amplification (WGA), enable anyone with basic equipment and know-how to produce practically unlimited amounts of in vitro synthesized (artificial) DNA with any desired genetic profile. This artificial DNA can then be applied to surfaces of objects or incorporated into genuine human tissues and planted in crime scenes. Here we show that the current forensic procedure fails to distinguish between such samples of blood, saliva, and touched surfaces with artificial DNA, and corresponding samples with in vivo generated (natural) DNA. Furthermore, genotyping of both artificial and natural samples with Profiler Plus® yielded full profiles with no anomalies.

Now, most people haven't perfected this technique, and don't have unlimited access to the equipment necessary to fake such DNA evidence. And yes, according to the NY Times, the authors of this paper want to sell their new technology/technique that can distinguish between real and fake DNA. But this is real, and they determine that not only can all current tests not distinguish between real and fake DNA, but that:

  • anyone with a biology major has the know-how to do this,
  • the equipment needed can be found in many standard biology research labs,
  • and only a small initial sample of DNA -- an eyelash, a single drop of blood, etc. -- is required to produce all the artificial DNA you like.

    The moral of this story? I'm going to be very careful next time before I piss off any biology majors...


More like this

I have the honor of answering the next installment of "Ask a ScienceBlogger". This time around the question is: Can a forensic scientist tell the difference between DNA from blood samples of identical twins? The short answer is "No." Identical, or monozygotic, twins occur when a fertilized egg…
I may be one of the few people in the world of TV watchers who has never seen a single episode of CSI: Whatever, a show featuring (I am told) "scientific" forensics work. Mrs. R., however, is fond of watching another show, Bones, which features a forensic anthropologist who works with a team that…
The transition from basic science to useful technology is so unpredictable. Look, for example, at DNA. Ten years ago, everybody assumed that the payoff of the genetic revolution would arrive in the form of gene therapy. We'd cure disease by tinkering with our genome. Well, that didn't work out so…
Last week, the European Court of Human Rights unanimously ruled that retaining DNA samples from innocent individuals in a national law enforcement databank violates human rights. The ruling is a direct blow to Britain's DNA databank, which holds samples and data for 7% of its citizens (4.5 million…

<insert obligatory xkcd link here>

Iono - they may have a point, but the fact that they want to sell the cure for the ill they've discovered makes them sound iffy.

I think that this is one of those things that while it could happen in some theoretical sense, in real life, an occurrence would be extremely rare.

Suddenly there is reasonable doubt for many convictions especially ones based solely on DNA evidence.

We've always had worries about DNA evidence, which, just like fingerprints, come with complete portability and utter timelessness.

Suppose you are our prime murder suspect, selected because you're the only one without a good alibi. Since you cannot prove you didn't do it, we need to prove you did do it.

During interminable questioning, among the tricks we pull is to hand you a cancelled envelope and ask you if you recognize the return address. You do not.

By the way, in Germany the cops were hunting for the weirdest serial killer ever, but they finally found the problem: a woman packing the DNA sampling kits was touching the swabs with her bare hands. It was her DNA that kept turning up at seemingly unrelated crime scenes.

Now we can convict you, since that envelope has your fingerprints and DNA on it, and we 'recovered' (cop-tricky-talk for 'found') at the crime scene. You cannot come up with a reasonable explanation for how your fingerprints came to be (ahem) found at the crime scene, so we are guaranteed a conviction. We will introduce the DNA in court, and the fingerprints, but not the envelope -- heh-heh. There is no way for you to prove we got the DNA and fingerprints on the envelope while we had you in custody.

You're a dead man walking. And we get credit for outstanding police work.

By Devlin Carnet (not verified) on 22 Aug 2009 #permalink

Silly me, taking crooked cops out of the equation.

It sounds like a case of selling a (non existent) problem along with a solution. Run some artificial DNA through the analyzer and look at the gel plate - I guarantee it's not going to look like anything you've ever got from a plant or animal. Not to mention that you need some pretty costly setup to produce fragments to interfere with the usual tests. In the absence of any connection to laboratories with such equipment, the DNA evidence stands. It's far easier to do other things to foul the tests to remove incriminating evidence rather than to plant evidence. There are numerous other issues in DNA analysis which simply dwarf this "fake DNA" non-issue and the labs deal with them every day. Remember that in the USA the key phrase is "reasonable doubt", not "far-fetched doubt".

By MadScientist (not verified) on 22 Aug 2009 #permalink


Suddenly there is reasonable doubt for many convictions especially ones based solely on DNA evidence.

Hardly reasonable. In fact, I completely fail to see how fake DNA is even a real concern. What can you do with fake DNA that can't already be done much more easily by planting a REAL sample where it doesn't belong? Seriously, if I want to frame you for my crime, I'll just steal your hairbrush and leave a few strands at the crime scene instead of mucking around in a lab.

This sounds like a company inventing a problem to sell the solution.


Good luck with that seeing how I have been bald for ten years!

This "fake" dna is still grown from a live sample, which you could have planted instead, for the same result. The question is, how long until you can grow truly fake dna based on a print-out of a dna pattern taking years ago? This invalidates the cold-case techniques of checking old samples against dna tests, for example taken routinely from prisoners, as the very fact that it's in a database would put the test result in question.

Sorry Ethan, I have to agree with your readers that support the 'too far fetch to worry about' opinion. While it is possible to copy DNA very easily, the knowledge and training required to perform the work is likely not going to be found in the people commonly convicted for murder or others crimes solved using DNA.

In the rare case that some bio lab rat researcher blows their top or loses their marbles and decides to become a psychopathic murder/rapist then we can be on the look out for fake DNA.

By Richard Helmich (not verified) on 24 Aug 2009 #permalink

@mu #9: It should be possible to work from an electronic copy of a profile. The "PCR" (polymerase chain reaction) is extremely effective. We also know what enzymes cleave DNA and where and what fragments are produced. All that information has been used in mapping genomes. In the past 20 years (or is it 30 already?) people have also synthesized amino acid chains (well, amino acids and the sugar-phostphate backbone) by growing them on polymer beads. After producing the exact amino acid sequence you want, you then break the chemical bond between your chain and the beads. So you can synthesize the many dozens of fragments you need to frame someone, run your small amount of synthesized fragments through your PCR equipment and get a much larger amount of the fragments you want. I don't know what you do next - put it in a spray bottle and spray all over the crime scene I guess. As already pointed out, why not just get genetic material from someone else and spread it around - it's much cheaper and it doesn't show up as a "fake".

By MadScientist (not verified) on 25 Aug 2009 #permalink