Wednesday, March 26, 2008

The limits of unlinked SNPs for learning about demography

The best way to learn about demography from population genetic data is to look at multiple unlinked regions (a common theme over at the evolgen blog). The distribution of frequencies in a populations of neutral alleles at SNPs (the site frequency spectrum), is informative about population history. For example an excess of low frequency mutations is consistent with recent population growth, as the increase in population size introduces new mutations but these mutations have not yet had time to drift to higher frequencies.

A number of papers have made use of the frequency spectrum of unlinked SNPs to learn about demography. A technical but elegant article by Myers et al shows that while informative about demography, the site frequency spectrum at unlinked SNPs can not help you chose between certain demographic histories. This is not a question of imperfect knowledge of the site frequency spectrum (which more data would solve) but because for any particular demographic model, as Myers et al formally show, there are a large family of demographic histories that can give rise to the same site frequency spectrum. They explained: 'Informally, changes in population size at some past time are canceled out by other changes in the opposite direction'. I think that this lack of information comes from the fact that each unlinked SNP only tells you about the placement of a single mutation on the genealogy of the population at that site, and over sites you learn about the expected amount of time in different parts of the genealogy (loosely worded, as there are in fact a set of genealogies). It is therefore perhaps not surprising in hindsight that these data are not sufficient to learn everything about population size changes.

This problem can be circumvented by using patterns of linkage disequilibrium within genomic regions to add additional information about the patterns of coalescent trees across the genome. I've often wondered along similar lines about how we can learn about population histories from population genetic data (even a whole genome's worth) and what the fundamental limits might be.

Myers S, Fefferman C, Patterson N.
Can one learn history from the allelic spectrum?
Theor Popul Biol. 2008 Jan 3

Sunday, March 23, 2008

Mutating males

A short but nice article by Doris Bachtrog, looking at whether there is a faster mutation rate in males compared to females in Drosophila. Studies of a number of vertebrates have shown a faster rate of substitution on the Y compared to the X at putatively neutral sites, thus suggesting a faster male mutation rate (as under neutrality the substitution rate is the mutation rate). This high rate of mutation in males is thought to be due to the larger number of cell divisions in the germ line of males compared females. Previous work has not seen this effect in Drosophila. Bachtrog used a set of orthologous genes on the recently formed neo-sex chromosomes of D. miranda (I discussed some of her work on this system in a previous post). The recent sex linkage of these chromosomes means that the genes on them have altered the proportion of generations they spend in the male and female germ line, making them an excellent system for looking at male and female mutation rates. To look at putatively neutral changes, Bachtrog looks at changes in short introns and at synonymous changes in genes (where the changes are classified by their direction with respect to codon bias). She finds a consistently lower rate of substitution at sites on the neo-X chromosome compared to the neo-Y, suggesting that in D miranda there is a higher rate of mutation in males.

I guess I'm still still slightly concerned that the higher rate of changes on the neo-Y could reflect relaxed constraint, but as the effect is seen across a range of different putatively neutral sites, a higher mutation rate seems the parsimonious conclusion. It would be interesting to know whether this higher male mutation rate is true of the entire Drosophila clade, or if is restricted to certain species due to differences in number of cell divisions. I wonder whether whole genome resequencing (or sequencing of a reduced representation) could be used on mutation accumulation lines to look into this. One issue with this experimental approach would be that generation time in the lab might differ from that in the wild, but it could offer a complementary way to look at this problem.

Bachtrog D. Evidence for male-driven evolution in Drosophila.
Mol Biol Evol. 2008 Apr;25(4):617-9

Friday, March 21, 2008

parasite induced mimicry in ants

Just a quick note to point people towards a paper (Yanoviak et al) that gives an amazing example of a parasite inducing mimicry in their host. The nematode parasite is transmitted to ant larvae via bird droppings, which the ant larvae are fed on. The parasite makes the ant's abdomen bright berry red, and also makes the ant hold its abdomen up in the air. This attracts birds to eat the berry-like parasite infected ants, completing the cycle.

There's a more extensive discussion of the article at This week in Evolution

Parasite-Induced Fruit Mimicry in a Tropical Canopy Ant
Yanoviak,Kaspari,Dudley, Poinar. Am Nat 2008. Vol. 171, pp. 536–544

Wednesday, March 19, 2008

Tan is the new black

There's a really interesting paper by Sean Carroll's group in Cell ( Jeong et al. ) on pigmentation differences between two closely related species of Drosophila (D. santomea and D yakuba). D santomea has a small range and is restricted high altitudes on the island of São Tomé, while Yakuba is more widespread and lives at lower altitudes on São Tomé. The species can hybridize, and form a natural hybrid zone, in fact the mtDNA has introgressed between species (Bachtrog et al. Llopart A et al). D. santomea unlike the rest of the Melanogaster clade lacks abdominal pigmentation. Carroll's group looks at one of the genes involved in this change in detail.

A couple of previous QTLs studies looking at abdominal pigmentation differences between these two species had identified a QTL close to a candidate gene tan on the X chromosome. While another small QTL maps near the Yellow gene also on the X chromosome. The authors show that Tan in combination with the Yellow gene produces the abdominal pigmentation in D. melanogaster. No coding differences are found between the tan gene of D santomea and D yakuba, suggesting that changes in the regulation are prime candidates for the difference in pigmentation. They then show that Yellow and Tan expression is present in the abdominal region of D Yakuba but absent in D Santomea. Replacing the X chromosome of D yakuba with that of D santomea removes the Tan expression pattern but not the yellow pattern, suggesting that the difference in Tan expression is controlled in cis (and yellow is controlled in trans). The authors identify a cis regulatory module for tan that controls the abdominal expression in D. melanogaster. They then show that this D. melanogater cis regulatory module can create the pigmentation pattern in male D. Santomea. A couple of changes have occured in the D. Santomea sequence in the regulatory module at otherwise conserved sites. The authors show that these sites are responsible for the reduced abdominal expression of tan.

At this point the authors decided to look at polymorphism and divergence data in D. santomea in this module. This is when the story gets really interesting. The authors I guess hoped to find the signal of a sweep in D. santomea around this region, and that all individuals would be fixed for the mutations that inactivated the cis module. What they found however was that the changes were not fixed in the population, but that there appear to be three distinct inactivation mutations at the cis module. They confirmed that the two newly discovered mutations (both deletions) removed the abdominal expression of tan, and so are likely to remove the pigmentation as well. Thus D. santomea has three different mutations at the same locus resulting in the same phenotype, which is pretty incredible case of parallel mutation. The authors argue that this is likely the result of selection rather than simply neutrality following relaxed constraint, and I find that pretty convincing. There is no observed polymorphism for pigmentation in D. santomea suggesting that the combined effect of these three mutations combined have removed pigmentation. It seems unlikely that this small mutational target (the module) experienced three neutral mutations that have essentially removed pigmentation.

I wonder if one of the mating choice QTLs (between D Santomea and D Yakuba) maps to the same location as the QTL that lead to the indentification of tan (I've not checked the co-ordinates of the QTLs in Moehring et al ). Interestingly, the yellow gene (the other pigmentation gene with reduced expression) seems to show a signal of introgression between D Santomea and D Yakuba (Llopart et al. ), I wonder if the
introgression at yellow gene has prevented the accumulation of strong cis mutations at yellow, meaning that its expression had to be reduced by a trans effect.

My only minor quibble with this otherwise great paper was the stridency about cis regulatory evolution. This paper is another really pretty example of cis regulatory evolution, but to my mind it in no way seals the debate about protein versus cis regulatory evolution ( Hoekstra and Coyne ). This case is another loss of function mutation, I would like to see more gain of function mutations in cis before making up my mind that cis regulatory evolution predominants. I also feel that the follow up of these cases is somewhat biased towards following up cis regulatory changes. The authors do not follow up the reduction in yellow expression which operates in trans, thus the paper has ignored a trans effect (which admittedly may be a cis effect at another upstream gene). Also the authors do not seem to show that the cis regulatory module of tan (a pleiotropic gene) is itself free of pleiotropic effects (a prerequisite for freeing up cis regulatory evolution).

Jeong S, Rebeiz M, Andolfatto P, Werner T, True J, Carroll SB. The evolution of gene regulation underlies a morphological difference between two Drosophila sister species. Cell. 2008 Mar 7;132(5):783-93.

Moehring AJ, Llopart A, Elwyn S, Coyne JA, Mackay TF. The genetic basis of prezygotic reproductive isolation between Drosophila santomea and D. yakuba due to mating preference.Genetics. 2006 May;173(1):215-23.

Llopart A, Lachaise D, Coyne JA.Multilocus analysis of introgression between two sympatric sister species of Drosophila: Drosophila yakuba and D. santomea. Genetics. 2005 Sep;171(1):197-210.

Bachtrog D, Thornton K, Clark A, Andolfatto P.
Extensive introgression of mitochondrial DNA relative to nuclear genes in the Drosophila yakuba species group.
Evolution Int J Org Evolution. 2006 Feb;60(2)

Friday, March 7, 2008

The descent of Y

One of the most obvious features of our genome, is the large difference between in the X and Y chromosomes. The Y chromosome is small compared to its partner the X, much of its DNA content is made up of repetitive DNA and it codes for few genes.

Much of the degeneration of the Y is thought to due to the lack of recombination on the Y chromosome. Recombination is thought to be initially suppressed between the homologous chromosomes around the sex determining locus, and then extend as sex-specific coadapted complexes of genes linked to the sex determining locus build up (which recombination would destroy). The lack of recombination means that the fate of mutations that occur on the Y chromosome are coupled: A beneficial mutation drags along deleterious mutations that occur on its background as it sweeps to fixation (hitchhiking). Deleterious variation also builds up as in the absence of recombination haplotypes lacking deleterious mutations can not be recreated once they are lost by genetic drift (Mullers rachet).

However, studying the formation of the human Y chromosome is difficult. The event happened millions of years ago, and while we can learn something about the event by looking over different mammalian species we are still very limited in what we can say. A different approach to learn about sex chromosomes is to look not at old sex chromosomes (where much of the action occurred long ago), but to study new sex chromosomes or new additions to sex chromosomes. This is one of the wonderful things about biology, no matter how strange the event, it is often occurring in multiple species independently as we speak. Thus if you want to learn about something that happened in the history of one species, looking for it happening currently in another is a great tactic. As sequencing and other resources become cheaper, this will make this approach in genetics even easier.

Young sex chromosomes have been identified in a number of species ( see here ), these are associated with locally suppressed recombination (around the sex determination locus), causing these regions to start to degenerate (this paper is also a good introduction to the evolutionary dynamics of sex chromosomes).

Another example of the degeneration of the Y chromosome due to lack of recombination, are fusions between autosomes and existing sex chromosomes. These fusions happen relatively often (fusions are a relatively common chromosomal abnormality in humans) and sometimes they survive and are fixed in the population. A number of fusions between sex chromosomes and autosomes are known in Drosophila. These vary in age, and so provide a good system for studying this problem. When an autosome becomes fused to the Y chromosome (a neo-Y) it ceased to recombine, as it is transmitted through males (which have no recombination in Drosophila ). The neo-X, which is the homolog of the neo-Y continues to experience recombination. A recent paper studies the decay of the neo-Y of Drosophia Miranda, this fusion is believed to have formed about a million years ago. The authors sequenced ~2.5Mb of neo-sex chromosome. They find that half the genes on the neo-Y are psuedogenized, due to the build up of deleterious mutations. While there counterparts on the neo-X are much more conserved. The Y chromosome is also rapidly accumulating transposable elements, around 20% of the neo-Y is transposable elements comapred to ~1% on the neo-X. Thus the degeneration of the neo-Y is incredibly rapid. This suggests that the homologous genes on the X chromosome will also have to evolve rapidly to cope with the loss of their partner. For example, if the neo-Y copy of the gene is non-functional, the copy on the neo-X will have to be up-regulated to compensate for this loss. Thus dosage compensation will have to evolve quickly on the neo-X. It will be really great to learn more about the evolution of sex chromosomes from these young sex chromosome systems, as more of them get sequenced.


Genomic degradation of a young Y chromosome in Drosophila miranda.
Bachtrog D, Hom E, Wong KM, Maside X, de Jong P. Genome Biology 2008

Steps in the evolution of heteromorphic sex chromosomes.
Charlesworth D, Charlesworth B, Marais G. Heredity. 2005

Sunday, March 2, 2008

right-handed snakes

A paper that was published last year, that gives a wonderful example co-evolution. I'm not sure if it got much coverage (but may well be mistaken), but it really is great.

Snails often have clockwise shells (which in itself, is a wonderful example of the evolution of asymmetry). The authors show that snakes that prey on these snails have evolved to have more teeth on their right mandible than there left mandible. The authors also show that these snakes have a lot easier time eating snails with clockwise shells, than those with anti-clockwise shells. I remember seeing a video of this at a conference I was attending, and thinking at the time that it was the coolest thing ever.

The mutation in snails, which causes them to flip the spiral of their shells is thought to be one of the only clear case of speciation caused by a single locus, as left-handed snails have a lot of trouble mating with right-handed snails (see here). But why do left-handed snails arise and spread if they can not breed with the right-handed snails? Perhaps predation by right-handed predators offers a mechanism that would favor the left-handed snail species.

Masaki Hoso, Takahiro Asami, Michio Hori. Right-handed snakes: convergent evolution of asymmetry for functional specialization. Biology Letters. 2007
Davison A, Chiba S, Barton NH, Clarke B. Speciation and gene flow between snails of opposite chirality. PLoS Biology 2005.