Can DNA Evidence Lie?
DNA evidence is foolproof, right? It depends how many fools are involved. Everyday Einstein puts DNA under the microscope.
Imagine one day that you're reading the classic fairy tale, Little Red Riding Hood. As you get to the climactic ending, you realize that this isn't the same version you read as a child. Instead you read something like this:
"Little Red Riding hood screamed. The woodcutter, hearing her screams ran into the house. Then Red ran out, saved from the wolf, all the way back home. Later that day, the local village constable arrested the woodcutter under suspicions that he had killed Red's grandmother."
Wait, what? Isn't the woodcutter the hero in this story? You can't believe that the local constable could even suspect this had happened. But then you turn the page and read:
"During the trial, DNA evidence was presented which linked the woodcutter to the crime. He was sentenced to life in prison."
Well if there was DNA evidence, surely that clinches it right? I mean DNA is...well DNA! Isn't it? But a small part of your brain is bothered by the fact that you don't actually know what DNA evidence means, so let's fix that. We’ll see exactly what DNA evidence is, how it's used, and when it can go horribly wrong.
As you likely remember from my episode on the Human Genome, our DNA is made up of a series of four different kinds of nucleotides, or bases, which scientists have given the names Guanine, Adenine, Thiamine, and Cytosine, or (GATC for short). There are 3.2 billion of these bases in the human genome. While a good portion of these are the same for every person on the planet, there are certain bits of DNA that are highly variable between individuals.
This variation comes because when your cells divide, they have to make copies of the DNA. Not all of those copies are perfect however, and sometimes parts of the DNA get repeated. For example one part of your DNA might be GCATATAT, but all of those AT's in a row confuse the DNA copying machinery in your cells, so the copy ends up with GCATATATATAT instead. While such errors can occasionally cause problems, most of the time they're no big deal.
Keep it in the Family
As you probably know, you have two sets of chromosomes. One set from your mother and one set from your father. That means that for a given bit of DNA, you'll have two copies. Each of these copies, called alleles, differs from person to person. Imagine for example that your Dad had version 1 and 2 of a certain allele, while your mom had version 3 and 4. This means that you'll either get version 1 and 3, 1 and 4, 2 and 3, or 2 and 4; a copy of one allele from each parent.
Since sequencing the entire human genome is a lengthy and expensive process, DNA evidence centers on looking at just certain sections of the DNA called markers, and comparing the alleles of various suspects with the alleles found in the evidence. Scientists have identified the genetic markers which provide the best variation between individuals, reducing the likelihood that two individuals would have the same alleles for all of the markers being analyzed.
Just the Facts, Ma’am
Looking at this explanation, it seems like DNA evidence is pretty foolproof. And it would be, if it were done consistently and properly in every case. Unfortunately many studies have found that this doesn't actually happen.
Let's go back to the Little Red Riding Hood case. Imagine that the woodland police come on the scene and find a dead wolf, a missing grandmother, and woodcutter holding a very suspicious-looking axe. Knowing that DNA evidence will be crucial to solving this case, they start gathering up these items from the scene of the crime. They put the axe, grandmother's dressing gown, and a sample of the wolf's saliva into a bag and take them down to the crime lab for analysis. The lab finds granny's DNA on the axe and concludes the woodcutter did it.
The critical question is, how did granny's DNA get on the woodcutter's axe? One possibility is that the woodcutter killed granny with the axe. However there are other possibilities as well:
The crime lab that analyzed the DNA evidence might have forgotten to properly clean their instruments, so granny's DNA wasn't originally on the axe at all.
Since the police stored the axe in the same location as granny's dressing grown, her DNA may have been transferred to the axe at that point.
Granny's DNA might not have been on the axe at all, instead a lab technician that was privy to the details of the case might have let his prejudice color the analysis.
The lab which analyzed the DNA evidence may have used a statistical analysis that isn't recommended by experts in the field.
Though our example is fictitious, all of these situations have occurred in real life. In addition, there are problems in being able to distinguish between the DNA samples of people that are very closely related, problems with DNA from certain cells degrading faster than others, confounding effects that can occur depending on the color of dye used in the analysis, and the way technicians choose to show the results of the analysis.
So what's our woodcutter to do? His best option at this point is to hire a good defense attorney who is familiar with DNA analysis and its limitations. In the long term, raised awareness of the limitations of DNA analysis could help prevent serious miscarriages of justice, but I will let Legal Lad to tackle that issue.
If you liked today’s episode, you can become a fan of Everyday Einstein on Facebook or follow me on Twitter. If you have a question that you’d like to see on a future episode, send me an email at firstname.lastname@example.org