sábado, 8 de fevereiro de 2014

Surfing genes

Hi! Today's subject is very intimately related to the previous post, but it's got a much more appealing name: surfing alleles!

Well, in the last post about genetic drift, we saw that this evolutionary force is nothing more than a lottery of genetic variants (or alleles) from one generation to the next. The consequences of this lottery are quite intriguing because we can observe that, eventually, we always get to a state of no diversity. IN average, this so-called fixation time (T) equals two times the number of chromosomes in the population (N): T=2*N. In summary, genetic drift is random (lottery) and destructive (reduces diversity)

Today, we'll have a look at a different context: range expansions. But what are range expansions? Or even, what are 'ranges', anyway?! So, a range is the area occupied by a given population. In other words, it is the spatial distribution of group of individuals. When this area (range) increases in size, almost always there is also an increase in the number of individuals, too. What is most important here, though, is that new areas become occupied. To simplify, we can imagine that the original population starts by being in only one determined area (or deme) and then began to occupy the surrounding demes (see image bellow). Each of these newly colonised demes can be named a new subpopulation, and these subpopulations put together form a metapopulation. Jargon aside, one population expands, takes over new territories and forms new subpopulations. So what?

So that every time one of these demes is occupied, we observe a phenomenon dubbed founder effect. This effect relates to the idea that, whenever a new area is occupied, not all individuals from the original population go there. I.e. each newly occupied deme takes but only a sample of the original population. If this sampling applies to individuals, it does apply as well to the genes within those individuals. This leaves a clear genetic signal of diversity loss in the new population. It's like there was an acceleration of the genetic drift effect.

Alright, but where is the surf of alleles? Well, the alleles hitch a ride on the wave of expansion, also in a random manner, but when it happens a very distinctive pattern is left behind. The figure bellow - extracted from Excoffier and Ray's 2008 paper in Trends in Ecology and Evolution - presents a very nice cartoon of what is the pattern left by this process:

What we see here is that the red allele, simply because it was in the right place at the right time (a), becoming the most frequent in the recently occupied areas (b,c). Were the green allele at the same spot, the same would probably have happened for it. This all happens because a range expansion is a series of founder-effect events. At every new deme that becomes occupied, only a sample of the previous deme's diversity is carried along. Soon enough, it becomes quite easy to track the fixation of one or other allele along the path of this expansion. This allele surfing has already been studied from many angles with simulations and also has many natural cases in which it seems to be the best explanation for observed patterns. However, maybe the most dramatic example of this process comes from experiments on bacteria, like the one run by Hallatscheck and colleagues in 2007 (PNAS), whose figure no. 1 is shown bellow:

In this figure, we see the development of a bacterial colony on a culture plate (A) that expands from it center to the edges (B), creating several patterns of fixation of red or the green alleles as highlighted (C). Overall, we can consider allele surfing to be spatial equivalent of genetic drift, which occurs in a single population through time. This idea was first presented by Slatkin and Excoffier in 2012 in the journal Genetics, and it makes a very nice reading for whoever is interested in more details of this story.

That's all for today. Allele surfing will definitely come back in more posts, once we have discussed natural selection. Come back soon to check it out and see you later!

quinta-feira, 6 de fevereiro de 2014

Drift, Genetic Drift

It was difficult to chose which would be the first post on this blog. There are so many different subjects that could be just as nice, but here goes. Today, we'll speak about genetic drift and, with it, start the series Evolutionary Forces.

So, first off, what are evolutionary forces? In a nutshell: forces that drive evolution. Biological evolution, that is. Well, what is biological evolution then? (We will call it just evolution anyway, OK?) Evolution is nothing more than change over time. Therefore evolution is not - as some people say - just a theory. It simply is the observed fact in the world that things tend to change, more and more, as time goes on. This applies to stars, planets, continents, rocks, and living things, being part of the universe, are no exception.

The mechanisms through which these changes happen in living beings, as well as the patterns left by them, is the subject of study for evolutionary biologists. Since the modern synthesis, four evolutionary forces have been recognised: (i) genetic drift, (ii) mutation, (iii) migration, (iv) natural selection. These can be analysed according to their effect on diversity. While mutation and migration create diversity (constructive), drift and selection are destructive forces, eliminating diversity either systematically or randomly.

Genetic drift is a random destructive evolutionary force. This means that drift leads to the loss of diversity in a random way. Let's analyse these two properties separately. Drift is a random process because any given allele (a genetic variant) in the population has the same chance of success in reproduction and survival. It is simply a lottery: whichever allele is lucky enough to be drawn will remain in the population.

In the figure bellow (from Wikipedia), we have an example. In the first jar we see a population with 50% red and 50% blue marbles. Now, imagine you (yes, you!) want to do a lottery draw to see what will be the next generation's composition. Now, there is important thing to observe: We can only fit 20 marbles inside a jar. What you do is to put your hand inside jar #1 and take one marble without looking (don't peek!)

OK, now put it back! And you may say: "Wait! Why?". Then I will say: "Very good question!". Well, here is why: what we are sampling are in fact genes that are passed to the next generation, not individuals. So, by chance, one lucky marble may have two kids while the other had none. That's what we're doing by putting the marble back in the "gene pool". For it to have a chance of being present two, three or more times in the following generation. So, what you do in reality is to take note of whatever marble color (allele) you took and put it back. When you're done with 20 drawings (the total amount of marbles we can fit in the jar), you go to your big stock of marbles (because you have a lot of them) and take enough marbles to fill the next generation's jar with the lucky alleles you have in your notes.

Now, if you repeat this process some times, you will always end up with one or the other allele (blue OR red) being the only kind you have in the entire population. And when you have only one kind of something you have zero diversity. Therefore a destructive process! If we had only drift happening in nature, there would be no diversity, and to be clear nearly no evolution.

There is one more detail, though. If you try to repeat this little experiment with much bigger jars with room for say 200 individuals, it will in average take you much longer to reach o monomorphic state (only red, or only blue). Actually, even our little example was quicker than the expected average. In a population of 20 marbles, the average time to have only one kind of marble is 40 generations. In a population with 200 marbles, this will take 400 generations! In a very very large population of 1 million marbles, it will take 2 million generations. Well, I guess you got it, right? It takes in average twice as much generations as you have of marbles (or chromosomes!) to reach a state with zero diversity.

If we want to make this whole story become more realistic (because we see diversity out there in nature!), at least one more thing is missing. We need new things to appear in the population. In the next article in the series, we'll look at mutation and how it brings diversity to populations. However in the next post - before looking at mutation - we'll see that alleles can surf! We'll investigate the intriguing phenomenon of allele surfing and see how this surfing is deeply related to genetic drift.

See you then!

quarta-feira, 5 de fevereiro de 2014


Hello! Please, be welcome. My name is Ricardo Kanitz, I'm finishing a PhD in population genetics and I am starting this blog as an attempt to tackle the popularization of population genetics or, well, Pop Genetics!

Here, I'll try to keep up with what's new and also classic in this fascinating field of knowledge. If you know something of population genetics, you may be imagining this will be a bunch of numbers and equations pilled up in between incomprehensible pieces of text. This is not the idea and - quite frankly - this is not my approach towards pop. gen. either. I actually hope that my intuition-based approach can make you more comfortable with this amazing science.

Also, this blog has a Portuguese-speaking companion here. In principle, everything you'll find here will also be available there, and vice-versa.