Science Sunday: telling time with DNA

By: Meradeth Houston on 10:01 AM

I'm back! Time to talk a little about some science stuff :) I probably have a bit more fun with these posts that those who stop by my blog, but oh well. Today I thought I'd talk a little about something called the Molecular Clock--or how we use DNA and mutations to tell time.

You may have heard science shows discuss how scientists have studied DNA from different populations or people and determined how old they are. It happens all the time. And, really, the underlying principle for how this kind of thing works is really quite simple. (I promise!)

What's been noticed is that mutations are always occurring in a really random fashion, but when you look at this randomness over a long enough time, a pattern starts to emerge. Basically, mutations (or differences in the A, T, C, & G's that make up a person's DNA in comparison with others) happen at a pretty regular rate. So while it's all still stochastic, we can say with a good amount of certainty that a new mutation will happen ever X years.

Okay, so if mutations happen at a pretty regular rate, all we need now is to know what the rate is. The way this can work (and there are several) is to look at two species or populations that we know exactly when they split apart. By comparing the differences that have happened since they split (assuming that they weren't there in the original, single population) with the known time, we have a rate!

Need an example? Say two groups of people separated into two populations 2000 years ago. If we look at the DNA differences between these two groups and see that there have been two mutations that occurred, then we could say that a new mutation arises every thousand years. (Yeah, world's most simple explanation & there are a few more complexities that make this more tricky, but I'm just going for broad overview here!)

Vincent Sarich and Allan Wilson were some of the first scientists to apply this fun little concept, noting that based on the mutational differences between a bunch of species (like horses, donkeys, mice, humans, and chimps) we could see how long ago different species split from one another. One of the really interesting things that came out of this was noting that humans and chimps actually split a lot more recently than had been previously thought, only about 6 million years ago (compared to 15, which was thought before).

This has been applied to a ton of other concepts (such as when people first arrived in the New World), and there is constant refinement of the rate of the clock going on, looking at things like generation length, areas of the genome that mutate more rapidly than others, etc. etc. While I personally think there are quite a few flaws  and hurdles to using the molecular clock accurately, it is an interesting tool and provides some great corroborating evidence!