I have to watch murder mysteries alone so that my muttering doesn’t disrupt the enjoyment of others. I’m willing to overlook that science procedures take place at an incredible speed in fancy surroundings – I mean, how boring would it be to watch actual PCR? – but I hate the implication that a DNA match means guilty.
So, repeat with me: DNA evidence can’t prove guilt. Why not?
DNA evidence is basically a matching game 1. You take DNA from your suspect and see if it matches to your victim or sample from the crime scene.
Who matches the crime scene sample?
Suspect 1 doesn’t match the sample. When someone has been exonerated by DNA results (e.g. David Milgaard in Canada), it is usually because the DNA doesn’t match. DNA evidence has only been available since 1989 (in Canada) and that was for special cases. Initial biological testing was done with limited blood types.
Suspect 2 matches the crime scene sample but does that mean she is guilty? No, there could be other reasons for her DNA to be present, depending on the crime and type of sample. For example, if there was a jewellery store robbery and her DNA was there, she could have been in the store before the heist.
Plus, there is a chance that Suspect 2 has the same DNA profile as someone else. A DNA profile is the overall pattern of ‘bands’ as seen above. But we hear all the time that our DNA makes us unique – how is this a concern? Well, we are unique but also 99.9% the same – confused yet? For forensic analysis, not every base (A,C,T, G) of our DNA is examined.
CODIS = Combined DNA Index System
This is the system that encompasses standards for testing and the database of information. Forensic DNA analysis focuses on 13 regions in the genome. These sections of DNA are known to be hyper variable (show lots of differences between people) and not originally associated with diseases.
Each region has many different possibilities so the chance of a match at each site can vary. Every time we add a region, the chance that someone else matches gets smaller. I’ll try to explain with some fake numbers:
| DNA region | Chance of that pattern | Probability of someone else matching |
| Area 1 | 1/10 | The odds are multiplied each time that you add an area |
| Add area 2 | 1/15 | 1/10 * 1/15 = 150 |
| Add area 3 | 1/10 | 1/10 * 1/15 * 1/10 = 1500 |
You can see how the numbers quickly go up as you add results from more regions. When you do all 13, the numbers get high. Plus. I’ve used nice even numbers here but sometimes the chance of someone’s pattern is higher than the numbers I’ve used. But there is always a chance that someone matches you.
Plus, our siblings also get half their DNA from our mom and half from our dad so the chance that your sibling matches is much higher. And if you have an identical twin, there is always someone who will match you.
Another way to think about DNA evidence
Thinking about small molecules is tough and we’re the same but we’re unique… So does this analogy work? Leaving your DNA behind at a crime scene is like having a photo of you in tiny little pieces. The detectives collect this information but can only see bits at a time. So they gather up the 13 pieces and compare them to their suspects (e.g. piece of ear, tip of toe, elbow, part of eye, etc.). If the tip of your toe doesn’t match, it isn’t your photo. But if you do match, do you believe that it is enough to guarantee you are guilty? Plus you could have left that photo there before the crime happened.
Just for fun, let’s try it. One of your neighbours has been spitting out their gum and you have stepped in it more times than you can count. So you decide to take the gum for DNA analysis. Somehow you convince all the neighbours to give their DNA for comparison (without warrants) and you can now try to match the results. If you think of DNA like the photo fragments, this should be easy.
I’ve collected 5 pieces of photo evidence for you.
Can you tell who the pieces match?
Neighbours
If you think you know, please leave your answer in the comments!
The pieces of evidence were taken from a puzzle of the offender’s image. If you want to try completing the puzzle of the offender, you can try it on this page.
——-
- I’ll discuss specifics of the procedures at another time. ↩
LisaM says
I like this post. Talking about the frequency of a particular allele in a large population is something that helps put the idea that DNA evidence proves innocence, not guilt. Some people/students often forget to consider the probability of getting a particular combination, so I like this example!
C.J. Benham says
DNA can’t prove innocence, either. The absence of someone’s DNA (or the presence of someone else’s) only proves that the first person didn’t leave any DNA at the scene. While this may strongly imply that the first person wasn’t present, it doesn’t prove it
genegeek says
True. I’m thinking of just the match itself. It is just a piece of evidence and should be examined along with all the other pieces.
Justus Nospam says
It’s the lady in purple.