Guest post by Andrew Hendry: "Pitchfork Science: Guppies, Stickleback, and Darwin’s Finches"

The following post is an invited personal guest post by Andrew Hendry at McGill University, who kindly invited me to write a guest post on "Eco-Evo-Evo-Eco"/Erik Svensson

By Andrew Hendry

I study Trinidadian guppies, threespine stickleback, and Darwin’s finches, surely 3 of the top 10 canonical vertebrate evolutionary biology “model” systems. I thus fall at one extreme (or is it three extremes?) on the “pick a model system and use it to answer my question” versus “develop a brand new system all my own” continuum. Many students and postdocs find themselves facing their own decisions about where to position themselves along this continuum. Should they take the shortcut of working with an established system so they don’t have to work out the simple details and can get right to addressing the big general questions? Or should they forge their own path and become an expert in something brand new? It might seem, based on the above listing, that I consciously took the first approach but the reality is something quite different. In truth, I used a “follow your nose” coincidence-and-serendipity approach to study system choice. I here trace my own personal history in these research areas before closing with some general thoughts on how to choose a study system.

Why I study salmon - a 16 year old me with a steelhead from our cabin (Kispiox River, BC, Canada).

I worked on salmon for my MSc and PhD, largely because I grew up with salmon fishing as my primary passion. Thus, I began studying salmon simply because I liked them and liked fishing for them. This led me to choose an institution (University of Washington - UW), department (School of Fisheries), and supervisor (Tom Quinn) who were ideally suited to immerse myself in salmon work. As my graduate work progressed, I very gradually became more and more interested in general questions in ecology, largely through exposure to the research of other people in the department. I even started subscribing to Ecology in addition to – of course – Fisheries. Yet my thinking remained salmon-centric: “what can ecology tell me about salmon”. Nothing wrong with that, of course. Then, when visiting home for Christmas in 1994, I received from my mother a book: “The Beak of the Finch” by Jonathan Weiner. When your Mom gives you a book for Christmas and you then spend the next week at home… well, you better read it.

The laboratory for my PhD - Lake Nerka, Wood River, Alaska.

The book was amazing. It described in wonderfully readable prose the research of Peter and Rosemary Grant on Darwin’s finches in the Galapagos Islands. What struck me the most, while reading beside the heater vent looking out at the blowing snow and -40 C weather (literally!), was Jonathan’s description of how the Grants had documented generation-by-generation rapid evolution of finch beaks in response to natural selection resulting from environmental change. Wow – you can actually study evolution in real time! It was my own eureka moment and, in short order, I became captivated by the idea. As soon as I got back to UW after Christmas, I went to the library and photocopied EVERY paper on Darwin’s finches (ah, libraries and photocopying – the good [and bad] old days). From then on, almost as though my brain had achieved an alternative stable state, my thinking was inverted to become “What can salmon tell me about evolution.” 

My MSc and PhD work focused on sockeye salmon - this one in Knutson Bay, Lake Iliamna, Alaska.

Salmon did tell me a lot about evolution. I even edited a book (Evolution Illuminated, with one of my evolutionary idols, Steve Stearns) about merging evolutionary theory and salmon research. However, when one starts focusing on a topic (evolution) rather than an organism (salmon), one starts to become irked by aspects of the organisms that are not optimal for addressing the topic. Most notably, it is very hard to do experiments with salmon unless you have lots of water, lots of space, and lots of time. So, when thinking about a postdoc, I started talking to folks about which systems might allow me to better address basic evolutionary questions. I ended up moving in two directions.

The laboratory for my first postdoc. For more than a month of glorious weather, I camped on a small island in a small lake (Mackie Lake) at the end of a 4-wheel drive road. Those are my mesocosms floating in the lake and projecting from the island.

The first was the University of British Columbia (UBC) – because I didn’t want to go too far from my girlfriend (now wife) who was still at UW. I visited UBC and went from prof to prof telling them of my interest in a basic evolutionary question – the balance between divergent selection and gene flow – and asking if they knew of a system that would be good for testing my ideas. Many great suggestions were made, but Rick Taylor insisted he had the perfect system: Misty lake-stream stickleback – and he was right. So I started working on stickleback not because they were a model system, but because someone suggested they would be well-suited for my question and because it let me stay reasonably near my sweetheart.

A threespine stickleback guarding his nest.

The second direction came about through a conversation with Ian Fleming, who suggested that I should work with David Reznick on guppies. I hadn’t even considered this possibility, but I knew a bit about the system (it is also described in The Beak of the Finch) and it seemed cool. So I went to UCR and met with David and talked about how we might use guppies to study the interaction between selection and gene flow. David said he would be happy to help me with this work but that he didn’t have any money for me – and so I offered to write a full NSF proposal. I was just gearing up to do so when I heard that I had received an NSERC (Canada) postdoctoral fellowship to work with Rick Taylor on the Misty system – so off I went to stickleback, leaving guppies behind.

My favorite wild guppies captured in my first year of sampling, 2002.

UBC was great, an outstanding place for nurturing interest and insight into general questions in evolutionary biology, but one must eventually move on. My next postdoc was the Darwin Fellowship (I applied because of the title) at the University of Massachusetts (UMASS) Amherst, working with BenLetcher on salmon again (hard to shake the habitat). While at UMASS, my guilt started building about telling David I would write an NSF grant and then not having done so, so I went ahead and wrote one, which got funded on the second shot (after bringing in my salmony lab-mate from Tom’s lab, Mike Kinnison). So my work on guppies eventually developed owing to guilt about not carrying through on something I said I would do.

The laboratory for our guppy work - here the Paria River, Trinidad

While at UMASS, my office happened to be near that of JeffPodos, who was working on Darwin’s finches. Near the end of my Darwin Fellowship, Jeff received an NSF Career grant and had money to burn – I mean invest. Jeff knew of my interests and asked if I wanted to come along to the Galapagos on the project (he recalls me asking – or perhaps begging – to come with him), and of course I immediately said yes. So my work on finches was simply a case of being in the right place at the right time. It was every bit as exciting as promised that cold winter back in 1994. Several years later, Jonathan Weiner called to talk about my salmon work and I was able to tell him how influential his book had been and how it actually brought me (without any plan) to work on finches.

In short, a large amount of coincidence and serendipity determined my choice of study systems. Once in each of the three systems, I became enamored with them and never left. I have now 25 papers on stickleback, 22 papers on guppies, and 11 papers on finches, and I have no intention of ever pulling back from any of these systems. I have also published 33 papers on salmon, and I continually look for new opportunities for additional work on them.

The laboratory for our finch work, presided over by a marine iguana.

Peter Grant once told me that, in conversation with Daniel Pauly at UBC, Dan told him that he (Peter) was a “point person” whereas he (Daniel) was a “line person”: a point person being someone who takes a single subject/system (finches) and looks at every aspect of their ecology and evolution, and a line person being some who takes a single subject (fisheries) and looks at it across many systems. I guess that makes me a pitch-fork person – trying to go into depth in three systems. Of course, this means that I can’t get too deep in any one system, much to my frustration. However, comparing and contrasting results from the three systems has proven fascinating. For instance, I study ecological speciation in all three systems with essentially the same methods (catching, banding/marking, measuring, recapturing, genotyping) focused on revealing the same processes (disruptive/divergent selection, adaptive divergence, assortative mating, gene flow). The similarities and differences in results obtained from the three systems has proved very instructive and motivational. In fact, my favorite research talk involves walking through a comparative story of ecological speciation in the three systems.

Perhaps my favorite finch photo.

Beyond how many systems one works in, I need to return to the question of working with model (developed) versus new (undeveloped) systems. As noted earlier, a benefit of working in a model system is that one doesn’t have to do as much background work (although every system is nowhere near as well-understood as the impression given by the literature), whereas a cost is that you are never known as the expert in that system (because the experts are the senior folks working on the same thing). The cost-benefit payoff is not easy to calculate and so the temptation for many students and postdocs is to spend a lot of time debating the pros and cons of the different approaches. I think all this angst is a mistake (or at least suboptimal) and that one should instead follow one’s nose (and Mom’s book recommendations). I think everyone should work on the systems and with the people that they find the most interesting and inspiring – not the systems that have the best-described genomes (as an example). These inspiring systems might be model systems or they might be new systems or both (I also have students work on non-model systems), but they are – most importantly – the systems that feel right at the time, not the systems that have been rationalized based on a logical calculation of optimal career advancement. It worked out fine for me (and many others) – although I am sure my colleagues would argue I could still use considerably more career advancement.

-----------------------------------------

Resources:

An interesting perspective by Joe Travis on question-based versus system-based science: Is it what we know or who we know? Choice of organism and robustness of inference in ecology and evolutionary biology

Fusing theory and data: a plea for help from Dan Bolnick

Posted by Erik Svensson


I got a request from Andrew Hendry, who in turn got a request from Dan Bolnick (University of Texas, at Austin) about the relationship between theory and data. There is a blog post at the  blog by Andrew Hendry's lab entitled "Fusing theory and data: a plea for help from Dan Bolnick". You might want to read the blog post by Dan and provide your comments and suggestions here, as I have done. This is an excellent example of the kind of scientific discussions that we could (and should) have on blogs, rather than keeping them at local coffee rooms. So take the opportunity to provide input and be part of the global research campus!

What do we need to know about speciation? Interesting TREE-review and discussion at Cell Press

Some of you might already have seen this, but I wish to draw your attention to an interesting TREE-review entitled "What do we need to know about speciation", published in Trends in Ecology & Evolution, and authored by Roger Butlin and co-workers in the FroSpecs-network (an ESF-funded research network focussed on speciation research). As you know, FroSpecs funds speciation conferences, meetings and small symposia, including one in Jyväskylä (Finland) next year, and one organized by us after the ISBE-meeting in Lund in August 2012 about behaviour, adaptive and non-adaptive speciation and ecological and non-ecological speciation.

The review by Butlin et al. aims to identify the most important future questions in speciation research in the coming years, and the article is also accompanied by an online discussion, where I was invited to participate, together with several other biologists, including Mike Ritchie, Maria Servedio and Andrew Hendry, to name some of our friends and colleagues. I encourage you to follow both the discussion and read the original article. In terms of speciation discussions, I would also like to recommend an interesting (albeit long!) blog post about "magic traits" on the research blog of Andrew Hendry.

Report from Uppsala: competition, ecological and non-ecological speciation

 Every now and then, one has to visit your enemy and competitor, as Richard Nixon realized in the early 1970'ties, when he visited The People's Republic of China, and shaked hand with communist leader Chairman Mao Zedong (see above). I imagine Nixon felt a bit unsecure when he, as an american, visited a traditional enemy on his home ground, almost like sticking your head in to the lion's den.

Uppsala and Lund Universities, being the oldest and most prestigious universities in Sweden, are often seen as competitors, but luckily we have not been close to armed conflict, like the US and China, and we are hopefully a bit closer to each other than Nixon and Mao. I was therefore honoured when I was offered to sit the thesis committé of Niclas Vallin, one of my colleage Anna Qvarnströms PhD-students, together with Prof. Andrew Hendry from McGill University (Canada). You probably remember that Andrew was also the opponent of my student Fabrice Eroukhmanoff in Lund, a couple of years ago, and then Anna Qvarnström was in the thesis-committe.

The current thesis by Vallin dealt with interspecific competition between flycatcher species on the island of Öland, and was a classical experimental field study which (almost refreshingly!) did not have a single chapter on molecular genetics, which is quite rare these days. Andrew writes a more thorough report about the thesis and its content on his research group blog "Eco-Evo-Evo-Eco".

On Thursday last week, Andrew and I also gave tandem talks at the Evolutionary Biology Centre (EBC) about the importance of ecological speciation, and its alternatives. We both took a critical look at ecological speciation, albeit from different angles, and Andrew writes more about it here.Briefly, Andrew questioned how often ecologically divergent selection leads to the completion of speciation, something which he calls "ecological non-speciation" , whereas I attacked ecological speciation with some examples of radiations which are unlikely to have speciated through ecological means and niche-based divergent selection, which we can call "non-ecological speciation".  

After long and scientific discussions over beers, wine and "Bäversnaps", Andrew and I agreed that we almost understood nothing, and that more research is clearly needed. I therefore would like to take the opportunity to, once again, advertise the ESF-workshop next year on non-adaptive and non-ecological speciation that will take place in Lund next year, on August 18 2012.

Lastly, I have to say I really enjoyed going to Uppsala (in spite of our historical antagonisms!), and to participate both in the thesis-committe of Niklas Vallin, and listen also to the thesis-defence of another PhD-student, Paolo Innocenti, who has worked on the transcriptomic consequences of sexual conflict in Drosophila. Interestingly, Paolo has worked both with Jessica Abbott and Tom Gosden, my two first PhD-students, so this is really a small world. And although Lund might still be the best university in Sweden, there is clearly room also for Uppsala, especially when they open up and collaborate with people from Lund.

Interesting blog about "Eco-evolutionary dynamics" by Andrew Hendry's lab

Those of you have met Andrew Hendry before, either in Vienna during the "Speciation"-meeting last week, or last year when he was the external opponent on the PhD-thesis by Fabrice Eroukhmanoff, might enjoy this blog from his research laboratory. His lab seem to have a similar attitude as ours when it comes to public outreach and using social media, which is nice. A general theme of their blog is "Eco-evolutionary dynamics", which fits well with the kind of research Hendry et al. have been pursuing in recent years, using sticklebacks, guppies and Galápagos finches as study organisms. Enjoy!

Greetings from "Speciation-meeting" in Vienna!

Together with several other colleagues from Lund, Sweden and other countries, our lab was well-represented at the first European Speciation Conference, organized by the Institute for Advanced System Analysis (IIASA) in Vienna (Austria). This three-day conference has gathered a number of researchers working on the problems of speciation, both theoretically and empirically. A list of talks from the conference can be found here.  

The first evening of the conference, we enjoyed nice Austrian food (LOTS of meat!) and good wine, and of course the company of many of our colleagues. On the picture above you can see how happy we are after tasting some great wine. From left to right you see Anna Runemark (Lund University), Fredrik Haas (currently at Oslo University), Erik Svensson (Lund University), Andrew Hendry (McGill University, Canada), Anna Qvarnström (Uppsala University) and Jörgen Ripa (Lund University).Although this time there was a Scandinavian bias at the table, we have also of course interacted and entertained ourselves with some other great colleagues, such as Maria Servedio as well as former McArthur student and legendary ecologist Mike Rosenzweig.

 Personally, I mostly enjoyed the talk by Daniel Bolnick (University of Texas at Austin), about the rarity of sympatric speciation in sticklebacks, which was somewhat heretical in a conference that has been so dominated by the "Adaptive Dynamics"-school, led by Ulf Dieckmann at IIASA, where the importance of sympatric speciation has been vastly exaggerated, in relation to its real importance in natural populations (in my personal opinion). If sympatric speciation was as common as these models of "evolutionary branching" indicate, there would essentially be a new species on every twig of a bush, which there clearly isn't. This very fact in itself suggests (at least to me) that constraints on sympatric speciation are likely to operate and be important, and that the asexual modelling approach in the adaptive dynamics school has underestimated the severity of recombination.

As a primarily empirically oriented evolutionary biologist, I see a major weakness of the Adaptive Dynamics-school in that their models are only weakly connected to empirical work and the parameters they include in their models are not as natural to estimate as the classical and well-established estimates typically  used by field evolutionary biologists that are derived from quantitative genetics (i. e. selection coefficients). In the absence of such transparent models with parameters defined in an empirically meaningful way, the jury is still out whether adaptive and sympatric speciation is really important in nature, or whether it is mainly a phenomena that gains more attention from theoreticians than it deserves from an empirical point of view of practicing naturalists and field biologists.

An exciting week with the phenotype in the centre of focus: Thesis nailing, lab-meeting (18 November) and dissertation

We have an exciting week in front of us, starting with the nailing of Fabrice Eroukhmanoffs PhD-thesis on Monday 16 October at 15.00. This ceremony will take place at "The Oak" in the bottom floor of the Ecology Building, and drinks will of course be served. Hope to see you all there!

On Wednesday (18 November), we will have our regular lab-meeting in "Darwin" at 10.00. Fabrice will show his Powerpoint-presentation to get some last feedback before the thesis defence on Friday 20 November. We will also discuss a recent paper by David Houle in the journal PNAS, where he suggests that time is now mature for the formation of a new scientific field: "Phenomics". After all the other "-omics"-revolutions in biology (genomics, transcriptomics and proteomics), Houle suggests that we should now return to the most interesting unit of evolution, and what made most of us interested in biology in the first place: The Phenotype. In spite of all the many advances in genomics and other reductionistic fields in molecular biology, our knowledge about how phenotypes evolve, and how they should be measured and quantified is still quite limited. Hopefully, this paper will open up our eyes for a bright future in the field of evolutionary ecology, and give some new idéas for research. You can download the paper here. If the links do not work, contact me or Anna Runemark (anna.runemark@zooekol.lu.se) and try to get a PDF from us instead. By, the way, do we have any "fika-volunteer" on Wednesday morning?

The exciting week does not end on Wednesday, luckily. On Thursday, Fabrice's thesis opponent, Professor Andrew Hendry from McGill University (Canada) will give a research seminar at 13.00 in the "Blue Hall" (note the time: it is one hour earlier than the "official" Thursday seminar which starts at 14.00). Hendry has done a lot of research on rapid evolutionary change in natural populations, gene flow, "eco-evolutionary dynamics" and ecological speciation. Thetitle of Andrew's talk on Thursday 19 November is:

"Ecological speciation (or the lack there-of) in sticklebacks, guppies and Darwin's finches"

Finally, this exciting week with the phenotype in focus will have a grand finale on Friday November 20 in the "Blue Hall" at 10.00, when Fabrice will defend his thesis. I hope as many as possible can and will join in to see Fabrice defending himself against Andrew Hendry, who is known to be a very critical and detail-oriented scientist. Most welcome!

New PhD-thesis in the lab: Fabrice Eroukhmanoff

I am pleased to announce that a new PhD-thesis will now be defended in our group: Fabrice Eroukhmanoff's Magnum Opus "The interplay Between Selection and Constraints on adaptive Divergence and Phenotypic Evolution".

This is the third Ph.D.-student that has finished his/her thesis in our lab, the previous two were Jessica Abbott (2006) and Tom Gosden (2008). You can find an abstract and more informaton about the thesis here. Well done Fabrice!

The thesis will be defended on Friday November 20 in the Blue Hall (Ecology Building). The external opponent will be Professor Andrew Hendry from McGill University Canada, and the thesis committé will consist of Professors Anna Qvarnström (Uppsala University), Karin Rengefors (Limnology, Lund University) and Janne S. Kotiaho (University of Jyväskylä, Finland). The thesis defence is open to everyone, and I encourage you to participate in this exciting event.

Relaxed selection and loss of non-beneficial traits

In the latest issue of Trends in Ecology & Evolution, there is a review and metaanalysis of the fascinating phenomen of trait loss after the disappearance of selection pressures maintaining the traits. Classical cases is the loss of vision among cave-dwelling fish or loss of flight ability or antipredator adaptations among birds and insects invading oceanic islands with few predators. This study is briefly reviewed at Science Daily, and one of the co-authors is Andrew P. Hendry, the external opponent of Fabrice Eroukhmanoff's Ph.D.-thesis on November 20 2009.

Fascinating questions to adress here is why do some traits disappear fast, while others take much longer time to decay? According to the results it seems as if two factors might be important in determining the speed by which traits are lost when no longer maintained by selection:

1/Traits that are energetically or nutrient-wise costly are more likely to disappear fast. Examples of such traits are the armour-plate reductions in marine sticklebacks, which disappear fast when these sticklebacks invade freshwater environments, where the minerals that are needed to produce these plates are scarce.

2/Traits that have a relatively simple genetic basis, and which are governed by one or a few loci are lost faster than traits governed by many traits. Examples of such traits include the loss of vision among cave-dwelling animals. Although many genes might influence vision, it might be sufficient with mutations in one or a few genes to cause blindness.

These interesting questions also apply to some of the study systems we are working in our lab, e. g. the Podarcis-lizards that Anna studies on the islets in Greece, where predators are few or the isopods that Fabrice have studied in Lake Krankesjön and Lake Tåkern which have invaded a new limnetic habitat (stonewort), where the isopoods seem to have evolved a suite of different anti-predator adaptations, perhaps as a response to a changed predator regime (invertebrates vs. fish) or perhaps even relaxed overall predation.