Carnival of Evolution #52

The Carnival of Evolution #52 is now up at the Genealogical World of Phylogenetic Networks. There are some interesting posts there from some excellent science writers.

Rapid speciation in starfish

Australian waters are extremely rich in starfish species. Indeed, a little over 15% of all known species of starfish occur in Australia. For at least two of these starfish, speciation occurred extraordinarily fast. At most, they became separated about 22 thousand years ago, but the best estimate for the timing of the split is about 6 thousand years ago.

We know that evolution can be very rapid (e.g. sticklebacks) and that sometimes this leads to speciation (e.g. cichlids). But, in these cases selection is probably acting on a small number of alleles that are already present in the population. What makes the starfish study so breathtaking is that there has been profound changes to life history in the two species, which likely involved selection on many morphological and physiological traits.

Puritz et al. looked at Cryptasterina pentagona and its sister species C. hystria. Like most starfish, C. pentagona has separate sexes and reproduces by 'broadcasting' sperm and eggs into the water column where fertilisation occurs. In stark contrast, C. hystria produces both sperm and eggs simultaneously, and it exclusively self-fertilises within its own body cavity. The embryos of C. pentagona develop in the plankton, while C. hystria broods its offspring within the gonad until they are ready to emerge as small starfish.

It takes an expert to distinguish Cryptasterina hystria (top) and C. pentagona (bottom) in the wild. In fact I've seen the bottom picture shown as C. hystria and C. pentagona, but I think I got it right (photo Jon Puritz).

Puritz et al. speculate that water temperature may have provided the selective pressure that favoured the evolution of the C. hystria life history. Viviparity, like that seen in C. hystria, has been documented in a number of other starfish species. And it is consistently associated with species that occur in colder water. The two Cryptasterina species are separated by about 375 kilometers, with C. pentagona in the warmer north and C. hystria to the cooler south.

The authors also argue that small population size may have selected for self-fertilisation. If there are so few individuals in the population that your gametes are unlikely to meet another individual's, it's better to fertilise your own than to not reproduce at all. It's expected that genetic variation in a population that self-fertilises should be very low. But, genetic variation in C. hystria is so low it suggests the whole species derived from very few individuals, perhaps just a single one.


The transition from broadcast spawning with planktonic larval development to self-fertilisation with larvae brooded within the gonad has occurred in another Cryptasterina species, C. pacifica. In the closely related genus Parvulastra, a similar transition has occurred too, but probably over 500 thousand years. This suggests that the genetic variation required for the dramatic shift in life history is widely present in the group of starfish to which the genera Cryptasterina and Parvulastra belong. But, the speed at which evolution has occurred is truly astonishing.

Parvulastra exigua, note its similarity to the Cryptasterina species (photo Museum Victoria).

Puritz JB, Keever CC, Addison JA, Byrne M, Hart MW, Grosberg RK, & Toonen RJ (2012). Extraordinarily rapid life-history divergence between Cryptasterina sea star species. Proceedings. Biological sciences / The Royal Society, 279 (1744), 3914-3922 PMID: 22810427

Creationists blow a fuse over chromosome 2

A while ago the gorilla genome was published. And it was pointed to by creationists as evidence that evolution is wrong (see my post here). Now the gorilla genome is fueling another spat between the forces of creationism and science. In particular what information that the gorilla genome can provide to help sort out the fusion on human chromosome 2.

Carl Zimmer explains the new evidence for fusion on his blog "The Loom". The discussion of the evidence for chromosome fusion is a prelude to a great story about creationist posturing and misdirection. Carl simply wanted to know what evidence there was to support their claim that the fusion of chromosome 2 couldn't have occurred since the split of our lineage from the chimpanzee lineage.

To hide the fact that there is no evidence to support this claim, they challenged him to a debate. When he turned them down for various good reasons and restated that he just wanted some evidence to support their claim, they crowed that he had folded. But, Larry Moran points out that they only like debates when they are on their turf and on their terms. Let's not forget that he didn't fold, he is still asking for evidence.

Let's also not forget that Carl has already outlined the case he would espouse in the debate. The creationists have not responded to the claims Carl made in his original post in any meaningful way. The fury with which they have confronted him is, to me, a little surprising. But, Larry Moran calls it "typical" and speculates about why Carl's request for evidence might have caused so much fuss.

***Update***

The creationists have now provided Carl with the evidence he was requesting. And, as expected, it's completely pathetic. Basically the evidence boils down to misrepresenting the findings of a paper that has been superseded by the more recent findings that Carl describes in his original post. On the upside, I've learnt some interesting things about chromosomes and genome evolution that I probably wouldn't have otherwise.

One fish, two fish, red fish, stickleback

Last week an interesting paper on evolution in sticklebacks, a widespread mostly marine fish, was published in Nature. Sticklebacks are fascinating because populations have repeatedly become established in freshwater at various times since the last ice age. Because of this they provide an amazing system in which the processes of adaptive evolution and speciation can be investigated in the wild.

The threespine stickleback, Gasterosteus aculeatus.

Evolution is often defined as the change in allele* frequencies with time. Observations of evolution in the wild, such as previous studies on sticklebacks, show that adaptation to novel environments can happen surprisingly quickly (just 13 generations in one documented case). Too rapidly, some argue, for adaptive alleles to have arisen after the novel environment is encountered. But, the adaptive alleles may already be present in the population at low frequency.

The left image shows the skeletal differences between marine and freshwater sticklebacks. On the right are preserved specimens stained red. Note the strong divergence in morphology between the marine and freshwater forms, which can arise in just 13 generations (all images David Kingsley).

If sticklebacks adapted to freshwater through the selection of alleles that arose after they encountered the new environment, then alleles within freshwater populations should be most genetically similar to the populations they diverged form. Conversely, if adaptation occurred through selection on already existing alleles, then freshwater populations should share similar alleles with each other. And the authors tested which of these possibilities was operating in sticklebacks by sequencing the entire genome of ten pairs of sticklebacks. Each pair came from the same area, but one member was the marine form while the other was the freshwater form.

They then examined the genomes for regions that were similar among the marine or among the freshwater fish. Using two different statistical approaches to determine similarities among regions they found 242 regions (0.5% of the total genome) were identified by either test and 147 (0.2% of the total genome) that were identified by both were divergent between marine and freshwater fish. That is, just by looking at those regions in a given fish you could be reasonably confident about whether it lived in marine or freshwater.

Next the authors looked at the genes contained within the regions that were divergent between marine and freshwater sticklebacks. They found that there was a significantly higher density of genes within the identified regions that the genome overall. Then, using the 64 regions that showed the strongest differentiation between marine and freshwater fish, they looked at whether the genes coded for proteins or had a regulatory function. Regulatory genes modify the function of other genes via proteins or RNA.

Regulatory genes were more common within the 64 strongly divergent regions. Just 11 regions (17%) contained coding genes, while 26 regions (41%) were regulatory. The other 27 regions (43%) contained both coding and regulatory sequences, but in these regions none of the changes to coding genes produced different proteins. This strongly suggests that these regions have a principally regulatory effect on trait expression. Thus, regulatory changes account for a much higher proportion of the differences between marine and freshwater sticklebacks.

So, this study is really cool for two reasons. It shows that rapid adaptive evolution to a novel environment can be achieved using the genetic variation present in the parent population. And it shows that the regulation of when and where coding genes are expressed largely accounts for the for the differences between marine and freshwater populations. As the authors acknowledge, the next step is to determine which traits are affected by these genetic differences between marine and freshwater populations.

*Alleles are variants of genes.

Reference (open access!):

Jones, F., Grabherr, M., Chan, Y., Russell, P., Mauceli, E., Johnson, J., Swofford, R., Pirun, M., Zody, M., White, S., Birney, E., Searle, S., Schmutz, J., Grimwood, J., Dickson, M., Myers, R., Miller, C., Summers, B., Knecht, A., Brady, S., Zhang, H., Pollen, A., Howes, T., Amemiya, C., Baldwin, J., Bloom, T., Jaffe, D., Nicol, R., Wilkinson, J., Lander, E., Di Palma, F., Lindblad-Toh, K., & Kingsley, D. (2012). The genomic basis of adaptive evolution in threespine sticklebacks Nature, 484 (7392), 55-61 DOI: 10.1038/nature10944

The gorilla genome

Recently the gorilla genome was published. It showed that 30% of the genome was closer to  humans or chimpanzees that chimps and humans were to each other. The creationists were delighted, for they though they thought that the 'evolutionists' had just provided them with evidence against common descent. They were wrong...

Nobody was surprised that they were wrong; it's a habit of their's. Similarly, nobody with a good understanding of evolutionary genetics was surprised that humans had some genes in common with gorillas that they didn't share with chimps. It was, indeed, expected. But, it provides an interesting lesson for some common misunderstandings of evolution.

There is a pervasive idea that humans are the pinnacle of evolution and that chimpanzees and gorillas are 'primitive'. But, in reality, chimps and gorillas have an equally long evolutionary history and have changed just as much as humans since our lineages split from one another. And one study has shown that more genes in the chimpanzee lineage have been under selection than genes in the human lineage since the split.

Another misunderstanding is that the genetic differences observed now, must have appeared at the time the lineage split. But, the divergence of two species is not instantaneous; the genetic divisions become deeper over time. Moreover, some genetic differences may have appeared long before the split and were lost, by chance, in the populations that gave rise to one lineage, but not the other. The tree diagram for individual genes may, therefore, look very different from the phylogenetic tree. This phenomenon is know as incomplete lineage sorting (ILS). 

The important thing for evolution is that on average the human genome is more similar to the chimpanzee genome than either are to the gorilla genome. And this is the case. I suspect that many of the creationists shouting about this latest paper disproving evolution know that this is the case. But, I also think they know that it is a technical and often misunderstood part of our evolutionary knowledge and are using it to spread doubt about evolution through misinformation.

To understand ILS, you have to think not just about genes, but about the populations and species that genes occur in. When there is more than one version of a gene, each version is known as an allele. In apes (and most animals) any given individual will have two copies of a gene, which may be different alleles. But, in a population there are likely to be alleles that aren't present in every individual. Similarly, in a species all possible alleles may not be present in every population.

As populations diverge to form species, some of the alleles present in the ancestor will be lost because they won't be present in both populations. Others, however, will be retained in both descendant species. Fast-forward to another division in one of the species and the same thing can happen. By chance some of the alleles present in the common ancestor to all three species will be present in only one of the recently diverged species and in the more distantly related group. If the creationists had read more than part of one sentence of the gorilla genome paper, they would have seen this process illustrated in the very first figure.

Figure 1 from Scally et al. (2012) showing the phylogenetic tree for humans (H), chimpanzees (C), gorillas (G) and orangutans (O) with an example of incomplete lineage sorting overlaid in grey. In the example, the branching of one gene (grey line) does not map with the average genetic distance between species (percentages at the bottom). If we were to look at the example gene only, we would infer a more recent common ancestor for chimps and gorillas than for humans and chips. And that's why we don't construct phylogenetic trees based on single genes. 

Further reading:
Scally, A., et al. (2012) Insights into hominid evolution from the gorilla genome sequence. Nature 483:169-175 


Bakewell, M. A., Shi, P., and Zhang, J. (2007) More genes underwent positive selection in chimpanzee evolution than in human evolution. Proceedings of the National Academy of Sciences 104 (18) 7489-7494