Friday, 29 May 2009

Genetic Bottlenecks

I'm not mucking with wolf systematics this week, not even a peek, as I'm focusing on another data set. I pushed out a manuscript to its completed first draft, and I'm starting to work on a second manuscript that unexpectedly grew out of the first. Re-reading one of my lab's papers turned out to be a good idea, as it's refreshed my thinking on Bottlenecks. And Bottlenecks are exactly what I want to discuss. .
No, Dave, not that kind of bottleneck

A bottleneck is when a population goes through an event that greatly reduces its size. I typically think of a bottleneck as being reduced to 10% of the pre-bottleneck population or less, but there's no hard, fast rules on what constitutes a bottleneck. Bottlenecks can occur from a variety of factors, but most frequently it's climate change that forces animals through them. Often, we find bottlenecks in species that were forced into refuges where they can survive, like Tahrs into the high mountains in a warm period, or Dall Sheep into foothills during glacial periods. Because the amount of space in the refugium (that's the fancy pants term for `refuge`) is limiting, the population is greatly reduced.

It's worth saying that these bottlenecks through refugia (pl. of refugium... I hate biologist-speak) are really important, evolutionarily. Because competition for resources is so high, animals tend to be come very well adapted to what they do. Anyone who doesn't doesn't leave behind as many kids, and are quickly outbred by their more successful cousins. But if bottlenecks are too tight, they can leave a species floundering on the shoals, because they've lost too much of their diversity, and too much of their capacity to adapt.

Here's a hand drawn figure!
This is what happens normally through time. The squiggly lines represent lineages, and the width of the tube represents the total population number. Since we frequently trace these with mtDNA (see my glossary to the right!), it represents the unbroken chain of female descendants. If you go back to your mother's great grandmother's grandmother, perhaps her sister had some children, but those children didn't have any children. Those would be a terminated lineage, represented by the little spurs that go out, but don't make it to the end of the tube. They stop early.

That's normal `lineage sorting,` as we call it. It doesn't require someone to have no children. It could be no female children, or no male children, or it could be that they didn't pass on the trait we're following. In a easy, simple example, if a tall female has children with a short male, but doesn't give any of the genes that made her tall to her children, they'll all be short. The lineage of descendants with the `tall` gene is broken, at that point. In our example, we start with three lineages, two of which are represented at the end. But each of the two lineages that made it has branched out, forming new spikes that persist. Some of them will go away, some of them will make up the successful families of tomorrow.

Here's what it looks like in a bottleneck
You can see that a lot of the lineages went away through the bottleneck. Only 1 of the 3 founding lines made it to the end. Compare this to the normal situation, where many more lineages persist. Because of the fewer surviving lineages, less diversity is passed on to the offspring, and the population at modern day is more alike itself (more monotypic) than those who didn't go through the bottle neck. By measuring the variability, we can look back and see which groups went through bottlenecks, and which didn't.

Warning: Dumb joke coming up.
So when I say I'm dating a bottleneck, it doesn't mean I'm going to a bar.

Some species that have gone through bottlenecks include Moose, Cheetahs, Galapagos Finches, some populations of Koalas, Northern Elephant Seals, and humans. At some point, the ancestral human population was reduced to around 15,000 people (possibly due to a volcano, I read?) which is part of why almost every human is closer related to every other human on earth than two random mallards, or caribou would be.


Alaskan Dave Down Under said...

Thanks for the info on that bottleneck! I'll try to find a supply of it down in Oz. Looking forward to tasting it.

LOL ;)

gpc said...

I've spent the last decade or so beating myself up for not studying science when I was in college or grad school. Now it seems the basis of everything I want to understand. So don't stop explaining!