Over at Dechronization, Liam Revell points to a recent report in Science that Microsoft has filed one or more patents [$-a] on the methods biologists use to reconstruct evolutionary relationships among living things.
The patent filing, by Stuart Ozer, claims invention of a variety of techniques already in wide use by systematists and evolutionary biologists – and (so far as I could tell) none of these inventions are original in quality. The whole patent filing can be read (at one’s own risk) in its entirety here, however I have also chosen a few select passages for reproduction, below.
Among the claims of invention in this patent filing, the author purports to originate:
“a method of generating biomolecular clustering patterns”
…
“counting evolutionary events in each of the identified plurality of positions at each identified node in the evolutionary tree”
…
“counting evolutionary events further includes: generating an event rate . . . wherein identifying related positions includes identifying related positions based on the event rate”
This is worrying to systematic biologists because they don’t typically worry about patenting new methods — in academia, the payoff from devising a new method is to have the paper in which you publish it cited by everyone who uses it. And, although it’s polite to ask the original author first, those methods are understood to be freely available for improvement and extension. The last thing anyone wants is to have to consult a patent attorney before publishing. The comments on Revell’s post reflect this perplexity and worry, and are well worth following.
Growing up in suburban Pennsylvania, where the most hazardous wildlife not extirpated from our woods is the occasional crazed whitetail deer, there was really only one danger I associated with the outdoors — ticks. Specifically, ticks carrying Lyme disease, a not-very-pleasant bacterial infection that attacks the joints, heart, and nervous system if left untreated. According to a paper released online early in Proceedings of the Royal Society B, my risk of picking up Lyme disease on an excursion into the woods behind my parents’ house may have depended on the diversity of bird and mammal species in those woods [$-a].
Sure, it looks like a giant rat, but that opossum is a walking death trap for disease-carrying ticks. Photos by ricmcarthur and jkirkhart35.
In a way, the ticks that carry Lyme disease are a threat to humans precisely because they don’t rely on us as a regular source for blood. Instead, they feed on a variety of mammals and birds, which allows them to maintain population densities high enough that a human wondering into a woodlot stands a good chance of picking up one or two of the little buggers.
But it turns out that not all of these non-human hosts are equally hospitable for ticks. The new paper’s authors, Keesey et al., caught a range of tick hosts — white-footed mice, eastern chipmunks, gray squirrels, opossums, veeries, and catbirds — and experimentally infested them with ticks. They found a huge range of tick success across the six host species: almost half of all ticks introduced onto mice were able to feed, while only 3.5% of ticks introduced onto opossums were. Most ticks that failed to feed disappeared — they were probably eaten when the host groomed itself.
The authors’ field surveys of ticks carried by these animals in the wild make the difference even more pronounced. Wild-caught opossums carried an average of almost 200 ticks — if that’s 3.5% of the ticks that try to feed on a opossum, then that means each opossum had attracted, and eaten, up to 5,500 ticks!
But the real impact of this result comes into focus in a mathematical model the authors develop to determine the effects of removing each of the six hosts from a woodland ecosystem. Removing intermediately-useful hosts like veeries or catbirds doesn’t have much effect on tick density. On the other hand, if you remove very tick-friendly hosts like the white-footed mice, tick populations plummet. And if you remove opossums, they increase dramatically. This is important because, the authors say, larger mammal species are the first to leave as patches of woodland are reduced to make way for human development — so an early effect of woodland fragmentation may be to reduce or eliminate opossums in that woodland, and boost the density of disease-bearing ticks.
This result goes a long way to fulfilling a proposal the authors made in a 2006 review article, that the diversity of alternative hosts for disease vectors like mosquitoes and ticks may shape the risk they pose to human populations [$-a]. It shows that, even in the relatively tame landscapes of suburbia, the way we humans manage what wildlife remains may have real consequences for our own well-being.
References
Keesing, F., Holt, R., & Ostfeld, R. (2006). Effects of species diversity on disease risk Ecology Letters, 9 (4), 485-98 DOI: 10.1111/j.1461-0248.2006.00885.x
Keesing, F., Brunner, J., Duerr, S., Killilea, M., LoGiudice, K., Schmidt, K., Vuong, H., & Ostfeld, R. (2009). Hosts as ecological traps for the vector of Lyme disease Proc. R. Soc. B, (online early) DOI: 10.1098/rspb.2009.1159
Introduced species can wreak havoc on the ecosystems they invade. But what happens after they’ve been established for centuries? A new study in the latest Proceedings of the Royal Society suggests that, in one case, an introduced species has actually become an important part of the native ecosystem — and helps protect native species from another invader [$-a].
Dingoes (above) control red foxes, which is good for native critters. Photos by ogwen and HyperViper.
The introduced species in question is the Australian dingo, the wild descendant of domestic dogs [$-a] that moved Down Under with the first humans to settle the continent. Today, 5,000 years after their introduction, dingoes are the largest predator in much of Australia, and they were a prominent part of the ecosystem encountered by European settlers. Europeans, like previous waves of human arrivals, brought their own domestic and semi-domestic animals — including red foxes, which prey on small native mammals.
The new study’s authors hypothesized that because dingoes reduce red fox activity both through direct predation and through competition for larger prey species, dingoes should reduce fox predation on the smallest native mammals. At the same time, dingoes prey on kangaroos, the largest herbivore in the Australian bush — and reducing kangaroo populations should increase grass cover, providing more habitat for small native mammals. When the authors compared study sites with dingoes present to sites where dingoes had been excluded to protect livestock, this is what they found: increased grass cover, and greater diversity of small native mammals where dingoes were present.
Recently a news article in Nature discussed ragamuffin earth [$-a] — the idea that human interference in nature has so dramatically changed natural systems that it may often be impossible to restore “pristine” ecological communities. In these cases, some ecologists say, conservation efforts might be better focused on how to maintain and improve the diversity and productivity of the novel ecosystems we’ve inadvertently created. It looks as though the dingo could be a poster child for exactly this approach.
References
Letnic, M., Koch, F., Gordon, C., Crowther, M., & Dickman, C. (2009). Keystone effects of an alien top-predator stem extinctions of native mammals Proc. R. Soc. B, 276 (1671), 3249-3256 DOI: 10.1098/rspb.2009.0574
Savolainen, P. (2004). A detailed picture of the origin of the Australian dingo, obtained from the study of mitochondrial DNA Proc. Nat. Acad. Sci. USA, 101 (33), 12387-90 DOI: 10.1073/pnas.0401814101
ResearchBlogging.org editor Dave Munger inaugurates his new column on the Seed Magazine website by drawing together RB-aggregated posts about milk — and giving very kind attention to my own recent post about a new study of lactation lactase persistence in European and African populations. (Thanks, Dave!) It looks like the new column will shape up to be another good way to keep abreast of the ever-expanding science blogosphere.
I am a super-powered mutant. For a given value of “super-powered” and “mutant,” anyway: I am an adult human who can drink milk. This is unusual among mammals, but as those (in retrospect, somewhat creepy) PSAs that used to run during my Saturday morning cartoons said, milk has a variety of nutritional benefits, if you can digest it. Which of these is behind the evolution of adult milk digestion in humans? According to a new paper in this week’s PLoS ONE, the benefit you get from drinking milk depends on where you live.
Originally, every human on Earth was lactose intolerant, like most mammals. That is, they lost the ability to digest lactose, the major sugar in milk, when their bodies stopped producing the necessary enzyme lactase after weaning. Then, some populations of humans domesticated milk-producing animals, and this seems to have generated strong natural selection [PDF] for a form of the lactase gene that remains active in adults.
In fact, milk-drinking populations in Europe and Africa have evolved “lactase persistence” independently [$-a]. This parallel evolution of a single trait motivates the new study by Gerbault et al. — drinking milk might have different advantages for African pastoralists and Northern European farmers. Milk has two major dietary benefits:
It’s generally nutritious as a source of protein and calories, and
Lactose can aid in calcium uptake in lieu of Vitamin D.
The main source of Vitamin D, for humans, is sun exposure — the UVB rays in natural sunlight stimulate production of the vitamin. In Africa, close to the equator, it’s easy to get plenty of direct sunlight; but in northern Europe, sunlight is less direct — so it’s harder to produce enough Vitamin D. (This is actually thought to be one reason for geographic differences in human skin color [$-a]: under lots of direct sunlight, dark skin is favored to minimize cancer risk; but under indirect sunlight, light skin is favored to allow more Vitamin D production.)
If the benefit of milk is calcium, not protein, then we would expect adult-active forms of the lactase-producing gene to be common in northern populations, and to decrease in frequency with decreasing latitude. This has been observed in a survey across Europe [$-a] — but while the north-south pattern supports the calcium-benefit hypothesis, it is not conclusive evidence. This is because the same pattern could arise without any natural selection acting on the gene — populations generally tend to be less genetically similar if they’re farther away from each other, a phenomenon called isolation by distance, or IBD [PDF]. In fact, Gerbault et al. find that the north-south pattern of genetic similarity is replicated in genes that probably aren’t under selection arising from life at high latitudes, suggesting that IBD, not selection, is responsible for the pattern in the lactase gene.
For a more conclusive test, Gerbault et al. developed computer simulations of the evolution of early European communities. By simulating populations’ evolution with different strengths of selection acting on the lactase gene, they could estimate how probable a particular value of selection was given the present-day frequency of lactase persistence in the real population — but also take into account the population genetic forces that create IBD. They found that in southern Europe, no natural selection was necessary to explain the present frequency of lactase persistence — but in the north, selection coefficients as high as 1.8% were needed. That is, in northern Europe, lactase persistence is so common that the simulations only produced the observed frequency when people who could not drink milk as adults had, on average, 1.8% fewer children than those who could.
In contrast to Europe, African communities don’t show the same gradual transition from frequent to rare lactase persistence, so IBD is less likely to explain the observed patterns. To explain the frequency of lactase persistence in African populations, the authors compared it to the frequency of pastoralism — and, finding a strong positive correlation, they concluded that lactase persistence evolved in Africa because it allowed shepherds to derive more nutrition from the animals they kept.
In short, widespread lactase persistence evolved in Africa because milk is a good source of protein; but it seems to have evolved in Europe because milk is a good source of bone-building calcium. Human populations on separate continents arrived at the same evolutionary solution, but for slightly different reasons.
Update, 18 October 2009: I’ve submitted this post to the NESCent competition for a travel award for the ScienceOnline 2010 conference in Durham, NC, January 14‐17th, 2010.
Diamond, J. (2005). Evolutionary biology: Geography and skin colour Nature, 435 (7040), 283-4 DOI: 10.1038/435283a
Gerbault, P., Moret, C., Currat, M., & Sanchez-Mazas, A. (2009). Impact of selection and demography on the diffusion of lactase persistence PLoS ONE, 4 (7) DOI: 10.1371/journal.pone.0006369
Ingram, C., Mulcare, C., Itan, Y., Thomas, M., & Swallow, D. (2008). Lactose digestion and the evolutionary genetics of lactase persistence Human Genetics, 124 (6), 579-91 DOI: 10.1007/s00439-008-0593-6
Swallow, D. (2003). Genetics of lactase persistence and lactose intolerance Annual Review of Genetics, 37 (1), 197-219 DOI: 10.1146/annurev.genet.37.110801.143820
Tishkoff, S., Reed, F., Ranciaro, A., Voight, B., Babbitt, C., Silverman, J., Powell, K., Mortensen, H., Hirbo, J., Osman, M., Ibrahim, M., Omar, S., Lema, G., Nyambo, T., Ghori, J., Bumpstead, S., Pritchard, J., Wray, G., & Deloukas, P. (2006). Convergent adaptation of human lactase persistence in Africa and Europe Nature Genetics, 39 (1), 31-40 DOI: 10.1038/ng1946
For a year between undergrad and graduate school, I interned with the ecologists at the Western Pennsylvania Conservancy. A fair bit of the work revolved around documenting the locations of rare plants or animals. But “rare” can be relative: common ravens (for instance) were hard to find in Pennsylvania, but much more abundant just over the border in New York. In Pennsylvania, striped maples are so abundant that they can interfere with the regrowth of logged woodland, but in Ohio they’re rare enough to be considered endangered.
This, of course, is because Pennsylvania’s human-drawn political boundaries straddle important ecological boundaries, like the transition from raven-friendly Circumboreal forests of New England to the raven-free forests to the south. And you might imagine that, even if there weren’t major topographic features within a particular set of human-created boundaries, there would still be enough climatic changes from one side to the other so that the species present at the southernmost edge of a flat, boring state like Kansas wouldn’t be entirely the same species present at the northernmost edge. That is, if you travel south to north across Kansas, you’ll probably pass a point where the lowest winter temperature becomes too low for some species.
Still, there are probably fewer species whose ranges end within the borders of Kansas than within Pennsylvania. As a paper in the latest Proceedings of the Royal Society attempts to show, this is for the intuitive reason I’ve tried to describe above: while Kansas is comparatively uniform, Pennsylvania is located at a transition between ecological regions. At such transitions, landscapes get complicated — and that complication, the paper’s authors say, is what helps create the boundaries of species distributions [$-a].
To make their point McInnes et al. examine the distribution of bird species across Africa. The broke the continent up into a grid, and scored each grid cell by the proportion of birds present in the square whose ranges had an edge within the cell, a measure they call “impermeability.” The impermeability of a given cell was strongly related to the number of habitat types represented in the cell — and these heterogeneous cells tended to be distributed along the boundaries of major ecological regions, like the Sahara Desert.
That is, species ranges tend to end in areas where desert is shading into grassland, or savanna into forest. Or, to put it another way, one group of species (birds) tend to have range edges that coincide with the boundaries of whole ecological communities. That’s an important observation — but it’s not exactly new. The entire concept of ecological communities arose because it’s obvious there are groups of living things that tend to occur together. McInnes et al. have quantified that observation, but they haven’t really explained it.
References
McInnes, L., Purvis, A., & Orme, C. (2009). Where do species’ geographic ranges stop and why? Landscape impermeability and the Afrotropical avifauna Proceedings of the Royal Society B: Biological Sciences, 276 (1670), 3063-70 DOI: 10.1098/rspb.2009.0656
A new Joshua tree study is just out in the current issue of New Phytologist, presenting an analysis of the environments occupied by the two different types of Joshua tree. The results demonstrate that the two tree types mostly grow in similar climatic conditions [PDF], which suggests that coevolution with its pollinators, not natural selection from differing environments, is responsible for the evolution of the two different tree types.
Hey, those Joshua trees look kinda different. Photo by jby.
The latest paper is a chapter from the dissertation of Will Godsoe, who just received his doctorate last week. It presents an analysis that sidesteps a fundamental problem with studying long-lived, specialized organisms — they’re hard use in fully controlled experiments. To determine whether the two types of Joshua tree really evolved as a result of coevolution with their pollinators, we’d like to be able to eliminate the alternative hypothesis that the two types evolved in response to different environmental conditions. Except for a small contact zone in central Nevada, each tree type occurs in a different part of the Mojave desert, and the two regions do have some broad-scale differences in when they receive precipitation.
Ideally, to determine whether two plants have different environmental needs, you just perform an experimental transplant, growing each plant in the other’s environment to see whether it fares as well as it does at home. This isn’t really possible with Joshua trees, which are pretty tricky to sprout from seeds (I’ve tried), and which, in any event, take something like twenty years to mature. So Will proposed to use niche modeling methods instead. Niche models are statistical descriptions of environments where an organism is known to live, often used to predict where it could live. To build niche models for each type of Joshua tree, Will assembled location data we’d collected over several field seasons in the Mojave, then spent another field trip driving around the desert some more to fill in the gaps — he wanted locations where Joshua trees were definitely growing and where they definitely weren’t, to fully “inform” the models.
Using the location data, it was possible to determine what kinds of climates each Joshua tree type tended to occupy by cross-referencing with existing climate databases, then fitting statistical models to the results. The models produced for each tree type could then be compared — and, for the most part, they’re similar. That is, if you collected seeds from one tree type, planted them where the other type grows, and waited around for a few decades to check the result, you’d probably find that it grew as well as it did in its home range.
So, if differing climates don’t explain the origin of the two types of Joshua tree, does that leave no other possibility but the pollinating moths? Not exactly — there are lots of environmental variables that weren’t available for Will’s niche models, for instance, or there could be a third, completely unknown factor. But this does make coevolution with the moths a more plausible explanation. In light of some of our very latest results — which should be going to press fairly soon — coevolution is looking like a better and better possibility.
Reference
Godsoe, W., Strand, E., Smith, C.I., Yoder, J.B., Esque, T., & Pellmyr, O. (2009). Divergence in an obligate mutualism is not explained by divergent climatic factors New Phytologist, 183 (3), 589-99 DOI: 10.1111/j.1469-8137.2009.02942.x
Godsoe, W., Yoder, J.B., Smith, C.I., & Pellmyr, O. (2008). Coevolution and divergence in the Joshua tree/yucca moth mutualism The American Naturalist, 171 (6), 816-23 DOI: 10.1086/587757
Smith, C.I., Godsoe, W., Tank, S., Yoder, J.B., & Pellmyr, O. (2008). Distinguishing coevolution from covicariance in an obligate pollination mutualism: Asynchronous divergence in Joshua tree and its pollinators. Evolution, 62 (10), 2676-87 DOI: 10.1111/j.1558-5646.2008.00500.x
Among the flowering plants, groups with flowers adapted to a narrower range of pollinators — the more specialized ones, like orchids or mints — tend to contain more species. Why? The classic hypothesis is that coevolution between plants and their pollinators leads to more pollinator-specialized plants, which are then more likely to become reproductively isolated, and eventually form separate species. However, I’ve just finished reading a review article that suggests an interesting alternative: that angiosperms may not be diverse because they’re specialized, but specialize because they’re diverse [$-a].
The review’s authors, Armbruster and Muchhala, first lay out a list of possible mechanisms connecting diversity and specialization. Three of them have specialization creating diversity, by (1) creating reproductive isolation, (2) enhancing isolation created by other forces, or (3) reducing extinction rates. Finally, there’s the possibility that diversity creates specialization, by (4) essentially forcing plants to divvy up the available pollinator community more and more finely.
Collinsia heterophylla, a member of a plant genus probably shaped by competition. Photo by Ken-Ichi.
The first two mechanisms are, as far as I’m concerned, contained within the classic specialization-creates-diversity hypothesis classically advanced by Verne Grant, that increased floral specialization makes it easier to form new species [$-a]. The third is a bit odd — generally, ecologists think that increased specialization means an increased, not a decreased, risk of extinction [$-a]. It’s intuitive that if you rely on fewer pollinator species, you can afford to lose fewer of them, and you have fewer opportunities to colonize new sites; so on the one hand, you’re at greater risk of local extinction, and on the other, you have difficulty establishing new populations. However, as Armbruster and Muchhala point out, this process should make more-specialized plant groups less diverse, which is the opposite of what we see.
The fourth hypothesis, that competition for pollinators causes greater to create greater specialization, leads to predictions that nicely differentiate it from the classic hypothesis: that hybridization between related flowering plants should be rare, and that plants should rarely occur in the same community as their closest evolutionary relatives. The first is important because it gives a reason to specialize on one or a few available pollinators — if a plant can’t reproduce with nearby relatives, all the pollen it exchanges with them represents wasted effort, and may actually interfere with pollen transfer from members of its own species. The second is a consequence of that process; plants are most likely to be able to hybridize with their evolutionary sisters, so successful speciation will usually require geographic or ecological isolation.
The authors then evaluate the evidence for these predictions in four plant genera with which they have prior experience: Dalechampia, Collinsia (pictured above), Burmeistera, and Stylidium. For these four groups, they find good support for the diversity-causes-specialization hypothesis — few natural, or even artificial hybrids, and few co-occurring sister species. To some degree, then, the new hypothesis is an effect of a researcher’s favorite study systems influencing their perspective on the broader picture of evolution. Armbruster and Muchhala give the same treatment to orchids, and find that for the most diverse angiosperm family, natural hybrids and co-occuring sister species are not rare. This ambiguity makes the review more interesting — it overturns the causation commonly inferred from the correlation between diversity and specialization, but it doesn’t make the mistake of sweepingly assuming the opposite instead.
Armbruster, W., & Muchhala, N. (2008). Associations between floral specialization and species diversity: Cause, effect, or correlation? Evolutionary Ecology, 23 (1), 159-79 DOI: 10.1007/s10682-008-9259-z
V. Grant (1949). Pollination systems as isolating mechanisms in angiosperms. Evolution, 3, 82-97
Johnson, S.D., & Steiner, K.E. (2000). Generalization versus specialization in plant pollination systems Trends in Ecology & Evolution, 15 (4), 140-3 DOI: 10.1016/S0169-5347(99)01811-X
Sargent, R. (2004). Floral symmetry affects speciation rates in angiosperms Proc. R. Soc. B, 271 (1539), 603-608 DOI: 10.1098/rspb.2003.2644
The Nature piece captures the concerns that came up when I first broached the subject of trying to increase the meetings’ online profile, especially the question of unwanted publicity: scientific meetings often serve as forums for presentation of work in progress — ideally, you’re hoping there will be people in the audience with interesting ideas for how to proceed — and presenters don’t necessarily want unfinished work broadcast all over the globe. Most obviously, there’s the (I would say slightly paranoid) fear of getting “scooped” because a rival reads about your work on a blog and kicks into high gear to publish first.
It’s not clear to me, however, that blogging really increases this risk; the people most interested in a given scientist’s work, and therefore most likely to work on similar things and potentially scoop her, are probably already in the live audience. And, furthermore, as the Nature piece points out, online coverage of work in progress could actually serve to establish priority in case of a real dispute. In any event, scientific societies are already adapting:
Conference organizers contacted by Nature had a wide range of policies on social networking. Many societies have banned digital photography in talks and poster sessions and some consider bloggers to be members of the media and subject them to certain reporting restrictions. …
Journals are also pondering how best to handle social networking at meetings. Nature generally supports social media tools, says Philip Campbell, Nature‘s editor-in-chief. And as long as it’s not a deliberate attempt to hype a new finding, he says that researchers should feel free to talk to colleagues who blog or twitter.
Blogging is increasingly recognized as a great way to communicate science to the public, and it seems likely we’ll see it become well-integrated into scientific meetings. Online coverage of Evolution 2009 was, I’d say, a good start. The page I set up to aggregate posts about the meetings drew 15 posts from 6 blogs, with most posting (12/16) occuring during the meetings. I’m still adding to that page as followup posts appear on participating blogs. If I had to organize that page again, I think I’d look for a better — maybe even automated — way to locate and link to posts. The volume wasn’t so much that I couldn’t handle it by monitoring a few RSS feeds myself; but I assume that blog coverage will increase in future years.
The FriendFeed I set up for the meetings drew less traffic than I’d hoped; 15 subscribers, and 74 contributions from various sources. I’ve broken down FF posts by topic in the graph on the right. In general, people used the FriendFeed about as I’d have predicted. They posted reactions to talks;
heard Mike Levine, Matt Rockman and Joe Thornton’s symposia at #E09… just brilliant
their own status during the meeting;
back from birding, time for some talks #E09
useful information;
visit the Systematic Biology exhibit for an amazing (free) Timetree of Life poster #E09
and, yes, they complained about the catering.
@mlabrum coffee seems to be a prohibited substance in Moscow. #E09
Fortunately conference coordinator Darrell Keim kept an eye on the feed, and was able to respond in some cases.
What we didn’t see much was back-and-forth discussion, as described at the 2008 meeting of the International Society for Computational Biology. A lot of this, I assume, is down to (1) the fact that computational biologists are more techy than your average evolutionary biologist, which contributed to (2) comparatively low subscription to the FriendFeed. I set the dedicated feed up just before the meeting, too, so there may not have been a lot of awareness that it was available until later. Traffic to the meeting website peaked the day before the meetings started (at 658 unique hits), and it takes a good lead-in to draw participation to something new like this. The official Twitter feed attracted 93 subscribers, and I linked to that from the main page very early on.
This meeting also saw the first webcast of any meeting activities — specifically, and appropriately, Eugenie Scott’s lecture on communicating science to the public. It wasn’t live, but it’s a start. Next year, it’d be great to see all the “flagship” lectures — the societies’ presidential addresses, maybe some of the symposia given by societal award recipients — put online.
This was all, as I have now said repeatedly, a good start. No previous Evolution meetings (to my knowledge) have aggregated related blog posts, or provided a near-real-time forum for reactions and discussion, or posted video from a major public lecture. I expect that the value of these resources will only increase as more people use them — we’ll have to see how things work out next year, in Portland, Oregon.
References
Batts, S., Anthis, N., & Smith, T. (2008). Advancing science through conversations: Bridging the gap between blogs and the academy PLoS Biology, 6 (9) DOI: 10.1371/journal.pbio.0060240
Saunders, N., Beltrão, P., Jensen, L., Jurczak, D., Krause, R., Kuhn, M., & Wu, S. (2009). Microblogging the ISMB: A new approach to conference reporting PLoS Computational Biology, 5 (1) DOI: 10.1371/journal.pcbi.1000263
