Question: Why Genetics?
Pardis Sabeti: Well, genetics is kind of like that perfect balance between math and biology so- It’s so-- It’s understanding the natural world and these principles going on in it but using a mathematical framework in order to understand the process so it’s for me perfect. It’s exactly the kinds of problems I like to think about.
Question: How have you changed the field?
Pardis Sabeti: Okay. Well, so I-- The people ask me this a lot and I like to-- about the sort of algorithm I developed and I- it’s nice but I like to always say that I put it in to a context, that it’s only one- it’s an incremental step. Many people before me and after me and working with me helped to contribute to it. I think it’s just a little bit about the way of looking at data so a lot of people for a long time-- Well, when you start going all the way back, even before Darwin laws people were thinking about what is evolution and how does it take place and just-- and the simple principle it’s all based on is if something is advantageous, if it somehow enhances the survival or reproductive success of its carriers then it’ll rapidly spread through the population. Right. So people who carry a certain genetic trait will be more likely to survive or reproduce so they’ll pass on their genes much faster, and so those genes that are good for you will spread through the population.
And so that’s a principle that’s been around for a long and a lot of other scientists have been thinking about ways of therefore looking for things that are very young and prevalent in populations that are likely to have spread through natural selection. And that’s really all I was doing is making an incremental step on how we study that, how do we detect things that are very prevalent in human populations but have rapidly spread through human populations. And so that’s what I was looking for.
Question: What are the implications of your work?
Pardis Sabeti: So it was just, like I said, working with lots of other people in my group so just to name a few David Reich, Eric Lander, David Altshuler. I was working in a community of people who were all thinking about looking at genetic variations, of how you might look at them and how you might understand them, and so reading lots of papers from other folks who were doing great work in that area I just looked at ways that you could basically go across the human genome and look at every variation, everything that’s variable between human populations.
So we call them sort of mutations that are polymorphic, multiple forms in the population, and I looked across the genome at every single one and said how common is it, how prevalent is it in the population, and then I used a framework looking at the- what’s going on in that region of the genome to date the mutation, to say how long ago it happened. And it’s just- it’s kind of like it’s sort of a process of decay. The longer a new variant has been around the more the background on which it exists begins to decay and so that’s what I was looking at as I just developed a way that you could measure how long something’s been around based on the background it exists on.
Recorded on: June 29, 2008