Wednesday, July 20, 2016

How to attend conferences


I am nearing the end of six conferences/symposia/workshop/seminar trips in six weeks. At one of the meetings (Canadian Society for Ecology and Evolution), I learned I was one of only four people (including Doug Morris, Miriam Richards, and Andrew Simons) to have attended each and every one of the previous meetings (11 of them). Now that the four of us realize this, we are in an escalating arms race to one day be the only person to have that distinction. All of this meeting-going has given me plenty of opportunity to reflect on the value of conferences and how to get the most out of them. A perfect topic to add to my ongoing series of “How To” posts for young scientists.

Stop 1. The Genomic Basis of Eco-Evolutionary Dynamics, Monte Verita, Switzerland.
A first major point is that attending conferences is extremely useful, and often fun, and you really should do so whenever opportunity and funds permit. One of the benefits is seeing the latest research, much of it unpublished, by researchers whose work you normally follow. More importantly, you often attend talks by people whose research you do not know, which is a great way to break out of your insular little clique of round-robin citations. Even more importantly, conferences are a great way to get new ideas, see new techniques, and find new research areas.

In reality, however, the above benefits could accrue from simply being very well read. By contrast, conferences provide one dramatic benefit – networking – that simply cannot be achieved by reading more literature. That is, the best thing you can do at a conference is simply to meet and talk to people, including old friends and colleagues, people whose work you follow but haven’t met, new people you didn’t know existed, and so on. Indeed, many of my most exciting and lasting collaborations have started through discussions at conferences.

Stop 2. Leibniz Institute of Freshwater Ecology and Inland Fisheries, Berlin. Then I took a side trip to Sanssouci, where I think I spotted Darwin on this ornate clock.

How best to network?

Go out of your way to talk to people. If it is someone whose work you know, introduce yourself. If it is a person standing in line beside you whose work you don’t know, introduce yourself. I have had many fascinating and influential conversations with people who were simply standing nearby. Of course, if the conversation isn’t interesting or is awkward or flagging, you can excuse yourself to go to the bathroom or get a drink or whatever.

Present a talk or give a poster. Most people at a conference will not know you or your work. By giving a talk or a poster, you can give people a teaser that will lead them to seek you out. Of course, giving a talk has other major benefits that stem from building your public-speaking skills, which is a great aid to career advancement. Give talks early in your career and give them as frequently as possible thereafter.

Attend social events. These informal settings are the best places to strike up relaxed (or intense) conversations with people. A great opening line is always “so, what do you work on?”

In addition to networking, a few other tips can help you get the most out of a conference:

Stop 3. Centre of Excellence in Eco- &Socio-Evolutionary Dynamics, Leuven, Belgium. (Here an earlier visit where I consider the co-evolution of beer and beer glasses.) 

Ask questions in talks! Perhaps paradoxically, the goal of asking questions isn’t to get information. Instead, most questions are intended to (and often do) reveal more about the questioner than they do about the speaker. As a result, asking questions is a great way to have other people be aware of you, which helps with networking. However, the main goal of asking questions is to make you think more critically about each talk. My buddy Mike Kinnison at one point challenged his students to try to ask one question in every talk they attended at a conference. Of course, that goal is probably impossible (and perhaps annoying to others) but a goal like that really changes the way you listen to a talk. Instead of passively recieving information, you begin an active internal dialog and assessment of what the person is saying and why. You pay much closer attention. You scrutinize each graph more carefully. You listen the specific words with more nuance. You seek holes and inconsistencies. All of this has the payoff of aiding and ingraining critical thinking – not just as an ability but as a goal. So, I would suggest writing down a single key question that you would ask for every talk you attend. Then try to ask them when you can.

Don’t attend too many talks. Everyone gets burned out eventually at conferences. The easiest way to avoid burnout is to not try to fill up your schedule with talk after talk after talk. Instead, leave yourself breaks to hang out, talk to people, etc. Also, instead of moving from room to room in an attempt to assemble your own schedule of talks, simply sit all the way through some of the sessions – even if the talk titles don’t seem especially relevant. I have seen many excellent talks simply by coincidence.

Don’t be on your computer or phone. While tempting, being on your computer inevitably disconnects you from the speaker and what they are saying. Of course, having your computer out is sometimes necessary – in fact, I have my computer out at the conference I am currently attending, largely because the 5 previous conferences have gotten me so far behind that I will otherwise miss key deadlines.

Stop 4. Joint Meeting of Ichthyologists and Herpetologists, New Orleans. Rich Lenski is everywhere it seems, even in the bathroom of a blues bar!

Which conferences should you attend?

Each of us usually has a conference that most closely matches our work, which we should therefore attend as often as possible. Sometimes this core conference can be two – one conceptual and one taxonomic. In addition, I highly recommend pseudo-random attendance at other meetings, which is an especially good way to get new ideas and meet new people. Most obviously, try to attend local meetings by societies that you would not normally be exposed to. And, not surprisingly, one can also pick conferences based on interesting locations.

Stop 5. Canadian Society for Ecology and Evolution, St. John's, NFLD. With a side trip to Cape Race - seen above.
Writing this, I started to wonder about my own patterns of conference attendance over the last 20 years, which I was able to piece together, focusing only on major core meetings, into the table below. Several points were of interest to me. First, my core conference changed through time with my core research interests. I started out focusing on fishy meetings (AFS, EEF, ASIH). As I became increasing interested in evolution, my go-to meeting became SSE/ASN, along with a new taxon-focused stickleback meeting that happens every three years. Most recently, I have focused on the Canadian Society for Ecology and Evolution (CSEE), partly because I have become increasingly interested in the INTERACTIONS between ecology and evolution and partly because I live in Canada and want to see ecology and evolution thrive in that context. The more haphazard meetings include an eclectic mix of fish meetings (ICBF), evolution meetings (ESEB), ecology meetings (ESA), genetics meetings (AGA), conservation meetings (SCB, The Wildlife Society), and general science meetings (AAAS). These other meetings were chosen largely for location (either near home or somewhere exotic) and – more recently – because I have been asked to speak in a symposium. Looking back, I think I have gained a ton from these meetings, mostly – as noted at the outset – the network of friends and colleagues that it has facilitated.



Problems

Conferences cost an annoying amount of money. Fortunately, most conferences and departments have funds available for student travel. Also, you can make them lower priced by attending nearby meetings, carpooling, staying in university dorms, and skipping the banquet (which is never worth the price unless it has a great party afterward). Some of these choices can also help to reduce your carbon impact.

Stop 6. Gordon Research Conference on Ocean Global Change Biology, Waterville, NH. With plenty of opportunities for great nearby hikes.


Monday, July 4, 2016

Special Issues: What are they good for?

I have helped to edit nine journal special issues and another two are in preparation. The published issues include 133 papers. I have spent a lot of time editing these special issues – and writing the introductory papers for them, and so it might be reasonable to ask: Is it worth it?


Eight special issues and one book I have helped to edit.

The main reasons I edit special issues are two-fold. First, I am usually asked to do so by a journal editor, a friend, or a colleague (sometimes all three being the same) – and I am a sucker for a personal invitation to help with something. Second, special issues have all the fun parts and none of the painful parts of editing: you get to structure the special issue the way you want, you get to invite the people you want to invite, acceptance rates are high, the papers tend to well cited, and it is generally a low-pressure exercise. However, to what extent can these personal assertions be supported by hard data?

Special issues come in two general forms: those associated with a particular symposium or those collecting papers by invitation outside of a symposium. The latter are perhaps more obviously beneficial in all the above respects. For instance, the 8 papers in the 2007 special issue of Functional Ecology that I helped to edit have been cited a 2048 times (Web of Science). However, conference special issues are perhaps less obviously beneficial and so warrant a formal analysis. In the most recent special issue that Katie Peichel and I helped to edit – the proceedings of the 8th International Conference onStickleback Behavior and Evolution – we decided to attempt such an analysis in our introductory paper. The rest of this post repeats parts of that paper, with permission from the journal (Evolutionary Ecology Research). If you are interested, more can be found in the paper itself.

Are conference special issues worthwhile?

Special issues resulting from conferences are often published in peer-reviewed journals, yet it might not be clear to what extent those issues are beneficial for journals, authors, or editors. To address this ambiguity, we here use citation statistics from the first seven stickleback special issues to quantitatively address two questions: (1) Should journals publish special issues? and (2) Should researchers publish their work in special issues? … Finally, we comment from personal experience how special issues are useful for the editors that shepherd them through the process.

Should journals publish special issues?

Using Web of Science, we obtained the mean per-year citation rate for papers published in the special issues and for papers published outside the special issues in the same journal and year (Fig. 1). In one instance (the first special issue in 1985), the journal seemingly did not benefit in that mean subsequent yearly citation rates were 0.55 for Behaviour papers in the special issue but 1.28 for Behaviour papers outside the special issue. In another instance (the most recent special issue in 2013), the journal seemingly did benefit in that subsequent citation rates were 1.86 for Evolutionary Ecology Research papers in the special issue but 0.43 for Evolutionary Ecology Research papers outside the special issue. In the other cases, however, mean citations for papers in the special issue were about the same as those for papers outside the special issue. We might thus safely conclude that – at the very least – journals do not suffer from publishing special issues based on conferences.


From the perspective of a journal with its eyes on prestige, the most pressing question might be: How do the above numbers compare to papers in our journal in the same year that were not published in the special issue? Extracting these numbers from Web of Science, we see that the top ten papers from the same journals/years have citation rates of 111-264, typically higher than the top ten papers from the stickleback issues. We then considered the most-cited paper inside versus outside the special issue for each journal/year (Fig. 2). The two numbers are very similar for the three most recent issues, which are those with the fewest total citations. For older special issues that have accrued more citations, the most cited paper is typically outside the special issue. Overall, however, results aren’t dramatically different for papers inside versus outside special issues, and they are likely biased against special issues because fewer papers are published there (145) than outside the special issue in the same journals/years (702 total).


Finally, we consider the “longevity” of influence for papers from special issues relative to other papers in the same journals/years. This analysis asks: Does the relative influence of papers inside versus outside special issues change with the age of the special issue? Figure 3 [here 2 of 4 panels are shown] shows the proportion of the total citations to papers published in each journal/year that were to papers in the special issue (normalized to proportion of papers published in the issue). Considerable variation occurs through time for each issue, presumably reflecting stochasticity; but the overall trend seems to be rather flat. That is, papers published in special issues have approximately the same longevity as papers published outside the special issue in the same journal/year.


We conclude that papers inside special issues are doing nearly as well, and sometimes better, than papers outside special issues in a given journal/year. More importantly, conference special issues mean that journals are supporting a collective scientific endeavor that brings researchers together in a shared arena to exchange ideas, which benefits the field even if it doesn’t directly benefit the journal. Surely this service to the scientific community is something that all journals should promote.

Should you publish in a special issue?

Authors typically select journals for submission according to some perceived optimal combination of exposure (and hence citation rates), prestige (and therefore hiring prospects), cost (depending on resources), accessibility (depending on philosophy), and probably of acceptance (depending on haste). These factors trade off with each other and even the optimal strategy might not pay off owing to inherent stochasticity in the system. How should special issues factor into an author’s decision? Typical benefits of such issues can include exposure (at least to people in the same field), cost (usually non-existent), and probability of acceptance (generally high). Typical – or at least perceived – detriments include prestige (sometimes assumed to be low) and exposure (to people outside the field). We can consider how some of these factors play out by examining citation rates.

We calculated the mean annual citation rate for stickleback papers inside special issues versus the entire literature in the same years, again excluding the above five journals (Fig. 4). In all cases, mean citation rates were lower for papers inside the special issue than in the literature as a whole, but the difference was often only modest. For instance, papers published inside special issues were cited 68% (mean of the yearly means) as often as typical stickleback papers from the same year. Given that most people probably do not submit their very best work to special issues, we speculate that papers in special issues actually punch well above their weight.


What about the longevity of a paper you publish in a special issue versus the literature as a whole? Here we calculated the total number of citations per year to the papers in each special issue relative to stickleback papers in general (i.e., in the special issue and elsewhere), excluding papers published in the five journals as noted above. For the older special issues (1985 and 1995), the relative importance of papers in the special issue started high and then stabilized to remain – ignoring some year-to-year fluctuations – relatively constant through time (Fig. 5 [2 of 4 panels are shown here]). For the subsequent special issues (2000 onward), the relative importance of stickleback papers increased through time, suggesting that papers published in special issues could have more staying power than papers published elsewhere.


We suggest that – if only from the perspective of citation rates – authors will probably want to target their very best empirical work to journals with a higher impact than those typically publishing conference special issues. However, it could be a better decision to target other solid work for special issues, where exposure to people in the same field will be higher, where citation rates can be as high as elsewhere, and where acceptance is more likely.

Should you edit a special issue?

First, editing a conference special issue affords an opportunity to read a diverse set of interesting papers – in essence, an editor gets a contemporary cross-section of work in a given field from a diversity of authors.

Second, the review process is much more efficient, collegial, and constructive than in other contexts. For instance, nearly all of the 58 people we invited to review papers for the present special issue agreed to do so, and nearly all of the reviews were very careful and constructive and submitted on time.

Third, making editorial decisions is usually much more enjoyable. Although not every paper submitted to a special issue is published there, acceptance rates are generally very high. With much of the anxiety about acceptance versus rejection out of the way, editing becomes a rewarding process of working with reviewers and authors to make the papers as good as they can be.

Fourth, conference special issues are important for the field, the community, and the continuance of the conference. For instance, they greatly facilitate scientific interactions by pointing authors to opportunities for cross-citation within the special issue, which can lead to new collaborations.

Yet we are stickleback biologists – so what would an editor not working with stickleback get out of a special issue? So let’s turn to Evolutionary Ecology Research Editor-In-Chief Mike Rosenzweig for his thoughts, which were sent to us when he reviewed this paper in manuscript form. “I would have added one important item to your list. Virtually all biological investigations gravitate towards a focus on one or a very few model systems. Each of these will be pursued with a doggedness and an encyclopedic comprehensiveness that cannot be accomplished without the focus. But how do we settle upon which systems will be our model systems? I believe special issues are likely to be part of the answer. For example, in my own case, I was well aware of investigations dealing with sticklebacks. But the special issues brought that awareness to a new level. I began to see more and more of the interlocking, fascinating and general problems that the stickleback crowd manages to engulf. Beautiful. I have no doubt that I would have failed to understand this without the special issues to focus my attention.”

Monday, June 20, 2016

Evolutionary consequences of sex-biased resistance

Males and females often differ in the average parasite loads that they carry, and explanations for this pattern abound. For example, differences in size, stress levels, hormones (see here), exposure to parasites, and investment in defense have been shown to explain variation in parasite loads between the sexes. Such sex-biased parasitism has important implications for host behaviour and population dynamics, effects that in turn can ripple through the community. Non-surprisingly then, sex-biased parasitism is a recurrent topic in ecological parasitology; and yet, the evolutionary consequences of sex-dependent variation in resistance - following changes in parasite selection – are generally ignored.

At the same time as I was wondering about those evolutionary consequences of sex-biased resistance, I found myself in a laboratory full of guppies (Poecilia reticulata) coming from a series of field introductions that I initially intended to use for exploring female evolution of defense under relaxed selection (i.e. parasite removal – see here). This seemed like an excellent opportunity first to use the males that I was breeding in the lab, and second to have a go at the more general, and understudied, subject of sex independent evolution. During the last two decades, awareness of the extent to which novel environments and changes in biotic interactions may lead to rapid evolution has increased dramatically to the point in which this is now a common consideration in basic ecological studies, conservation efforts and management plans. Yet, one potential limitation of most studies is that they either focus on one sex or conflate both sexes as if they ought to experience the same level of selection and show the same response to selection. In all fairness, such an assumption might be acceptable in some instances, but should be carefully considered for sexually dimorphic traits or when the potential for sex-biased selection exists. And, of course, one such trait that may show such sex-independent responses to the same environmental change is at the core of my research interests: defence against parasites.


The extent to which males and females might evolve differently in response to the same environmental change is not straightforward, and thus worthy of exploration. On one extreme of a continuum, we might expect the sexes to show similar evolutionary responses given that they might experience a similar environment and share most of their genetic background. On the other extreme of the same continuum, we can expect the sexes to show different responses given that they could experience the same environment in different ways and differ in some important regions of their genome – as is well documented for myriads of behavioural, morphological and physiological sexually-dimorphic traits. Therefore, my collaborators and I decided to test whether male and female guppies  showing sexually dimorphic resistance to a common and deleterious ectoparasite (Gyrodactylus turnbulli) evolved following similar or divergent trajectories, after the experimental removal of said parasite in four replicate populations (Dargent et al. 2016).

First, we confirmed that resistance to Gyrodactylus differed between males and females from the (parasite-present) ancestral population used to seed the (parasite-free) introductions. Indeed, males had higher resistance than females, and this effect was not caused by males being smaller in size – and thus providing less surface area for Gyrodactylus to grow – than females. Second, as we showed previously (here), we found that females in the four Gyrodactylus-released introductions rapidly (4 and 8 generations) and repeatedly evolved increased resistance to the parasite. Interestingly, males did not evolve increased or decreased resistance in those same four and eight generations. Surprisingly, all four female populations shared the same evolutionary trajectories in resistance trait-space (i.e. they evolved in parallel) and towards the position of the ancestral male traits.

Although a potential argument could be that males evolve much more slowly than females, this does not seem to be the case here; males from these same populations have shown rapid evolution of other traits (e.g. colour – see here). Additionally, the explanations that we had outlined for female increased evolution of resistance (here) – that it was a pleitropic by-product of evolution in response to release from predation - does not hold well for males. Males in the source population experience stronger selection by predators than females, and therefore we would have expected to see an even larger increase in resistance. Having no clear selective explanations for the sex-specific evolution of resistance we consider the idea of stronger evolutionary constrains in males than in females. Put simply, either ancestral selection on resistance was stronger for males and depleted much of the available genetic variance, and thus constrained further evolution, or alternatively, the costs of increasing resistance are nonlinear making progress towards ever higher resistance progressively more costly.
Evolutionary trajectories of guppies in resistance trait-space (parasite intrinsic rate of increase (r) and peak load). (a) Females from the introduced populations (filled symbols) evolved increased resistance to Gyrodactylus spp. relative to ancestral females from the source population (S), while males  did not(empty symbols). Ellipses represent 75% data spread for females (b) and males (c).
Studies of sex-biased parasitism recognize that behavioural and physiological differences between the sexes can lead to divergent parasite loads, and nonetheless, these studies ignore the potential effects of sex-specific evolution on those traits that influence host infection levels. Our results show that sex-biased resistance is a highly dynamic character, and help clarify why it has been so challenging to establish general patterns and mechanisms of sex-biased parasitism.

The paper:
Dargent, F., Rolshausen, G., Hendry, A.P., Scott, M.E. & Fussmann, G.F. (2016). Parting ways: Parasite release in nature leads to sex specific evolution of defense. Journal of Evolutionary Biology, 29(1), 23-34.

Thursday, June 16, 2016

EcoEvo Deathmatch: Behavioral types versus Eco types

Some time ago, I wrote a post called “High Enthusiasm and Low R-Squared” in which I commented on how some research subjects seem to garner interest far greater than their real importance to ecology and evolution. One example I gave was genes of large effect, which are all the rage in scientific tabloids yet probably contribute to only a minor fraction of overall adaptation. Another example was biodiversity-ecosystem function relationships, which often have low explanatory power. That is, for a given biodiversity level, the range of ecosystem function is very large – even within a single experiment. A third example was parallel (or convergent) evolution, where instead most adaptation seems to be non-parallel (and non-convergent).

From my forthcoming book "Eco-Evolutionary Dynamics"
The fourth example I gave was so called “behavioral types” or “personalities” (and the related idea of behavioral “syndromes”), for which I argued that – in reality – behavior at any one time (or context) is usually not very predictive of behavior at another time or context. I didn’t mean to suggest that behavioral types weren’t interesting and, in fact, my former postdoc Lisa Jacquin just published in the Journal of Evolutionary Biology our cool study of how behavioral types evolve in Trinidadian guppies in  response to different predation and parasitism regimes. (Although a reviewer made us excise most mentions of “personality” from the MS.) Rather, the point of my original post was simply that behavioral types might be overblown with regard how much research emphasis was placed on them.

Bold versus shy for guppies from different predation and parasitism regimes - from Jacquin et al. (2016 - J Evol Biol).
I am currently toward the end of a two-week trip to Europe that included stops in southern Switzerland, Zurich, Berlin, and Leuven (Belgium). After the first stop, which was for a conference/workshop on the “Genomic basis of eco-evolutionary dynamics,” I wrote a post that revised my original criticism of one of the above areas – genes of large effect. While I remain confident that most adaptation is the result of genes of small-to-modest effect, I am now also interested in the possibility that some specific genes might have reasonably large effects on eco-evolutionary dynamics. We might call them “keystone genes” in echo of Bob Paine’s keystone species idea. (Sadly Bob recently passed away.)

My second visit on the trip was to the IGB (Leibniz Institute of Freshwater Ecology and Inland Fisheries), where I was hosted by Robert Arlinghaus. A series of discussions on that visit have motivated me to revisit another of my suggested areas of “high enthusiasm and low r-squared.” A number of people at the IGB study behavioral types and some also examine the influence of those types on ecological processes. Yet- to expand my earlier criticism – behavioral types might not have much influence on ecological processes because behavior is quite variable to begin with (i.e., “types” are not really that consistent) and individual-level behavior might or might not have much influence on ecological function. To address these concerns we need to calculate the effect size of behavioral types, ideally in relation to how the same ecological parameter is influenced by some other causal force that we already know is important. So what we need is an experiment that asks about the effect size of behavioral types in relation to other drivers of ecological function – and why not “ecotypes”?

Ecotypes are, classically, populations adapted to particular environments, such as benthic versus limnetic feeding environments for fishes, different soil types for plants, high-predation versus low-predation environments, and so on. A lot of work on fish (whitefish, stickleback, guppies, alewives) has shown that these ecotypes differ in their influence on various community and ecosystem level ecological processes. So why not implement an experiment explicitly comparing the ecological influence of different ecotypes (e.g., fish obtained from populations adapted to different environments) to the ecological influence of different behavioral types (e.g., fish within those populations that are either bold or shy). Although this particular comparison would certainly be interesting, it is unfair in one respect. The “ecotype” effect is between populations, where evolutionary divergence is possible, whereas the behavioral type effect is within populations, where evolutionary divergence is more difficult.

The ecological effects of some fish ecotypes - also from my book.
Fortunately, we do have a “fair” and appropriate comparison to make. In addition to situations where benthic versus limnetic ecotypes are separate populations, such as in different lakes and sometimes even within lakes, many fish populations also show continuous quantitative variation among individuals along a continuum from limnetic to benthic. That is, within any given population in a given lake, some individuals will be specialized for limnetic feeding and others for benthic feeding. Using such a population, one could perform a mesocosm common “gardening” experiment crossing behavioral type (presumably assayed before the experiment) with ecotype (perhaps based on capture location – inshore versus offshore – or on characteristic morphology or coloration). One could then assess the relative importance of these two factors for the usual ecological parameters, such as zooplankton abundance, water clarity, DOC, benthic invertebrate communities, decomposition rates, and so on.

Beyond the just-noted benefit of allowing a direct comparison between the effects of behavioral type and ecotype, this experiment has another advantage – it can consider interactions between the two levels of variation. For instance, the effects of a given behavioral type (e.g., bold or shy) might be evident only for a particular ecotype, and so an experiment crossing the two levels of variation has the potential to increase one’s ability to detect the effects of either. Of course, the effect of a given behavioral type or ecotype on ecological variables likely also depends on the testing environment: bold versus shy might only matter for benthic fish in benthic environments. So one would ideally cross behavioral type with ecotype with testing environment in a fully crossed design. In addition, if one is to make general statements about the effects of ecotype or behavioral type, one would want to do the experiment for at least two independent populations – that is, testing the parallelism across evolutionary replicates of interactions between behavioral type, ecotype, and testing environment shaping ecological function. Yes, I realize this is a massive undertaking but I think we should at least consider the ideal experiment before then chopping it down to a more manageable subset.

Behavioral types and ecotypes might be correlated within populations (benthic fish might be more shy) and so separating the two effects might be difficult. Yet it remains critical. Imagine that behavioral types are closely correlated with ecotypes, such as when foraging environment or predator environment leads to the evolution of different behavior types – or vice versa. In such cases, an experiment focusing on only one or the other axis of variation would be unable to determine the true causality – because the two axes (behavioral type and ecotype) are closely correlated. That is, the apparent differences in ecological effects between two behavioral types might arise simply because behavioral type is correlated with (other) aspects of ecotype – and those other aspects are what drives the ecological effects. In this case, behavioral type is not the causal factor despite appearing so in the experiment. Thus, it seems most profitable to first examine the association between behavioral type and ecotype (within and among populations) and then – through careful selection of individuals that break the mold – that is, that cross the two factors to the extent possible.


I hope that this post will inspire someone toward such an experiment. If behavioral types prove to be as important as ecotypes for eco-evolutionary dynamics, then I will happily extol their virtues in all future attempts opportunities. More generally, an experiment such as that suggested here would help show other evolutionary biologists and ecologists that the study of behavioral types is important for our understanding of eco-evolutionary dynamics.

Saturday, June 11, 2016

Keystone Genes?

(The title of this post is new as of June 15, 2016, in honor of Bob Paine, who just passed away.)

While writing my book on Eco-Evolutionary Dymamics, I wanted a chapter on the genetic/genomic underpinnings of the interactions between ecology and evolution. About the time I finished the chapter, I received an invitation to submit a paper to Heredity, and so I converted the book chapter into a paper (Hendry 2013).


In the paper, I suggested that “The genetics and genomics of eco-evolutionary dynamics will be – to a large extent – the genetics and genomics of phenotypic traits” (more about this below) and then concluded (from the abstract):

(1) Considerable additive genetic variance is present for most traits in most populations.

(2) Trait correlations do not consistently oppose selection.

(3) Adaptive differences between populations often involve dominance and epistasis.

(4) Most adaptation is the result of genes of small-to-modest effect,

although (5) some genes certainly have larger effects than the others.

(6) Adaptation by independent lineages to similar environments is mostly driven by different alleles/genes.

(7) Adaptation to new environments is mostly driven by standing genetic variation, although new mutations can be important in some instances.

(8) Adaptation is driven by both structural and regulatory genetic variation, with recent studies emphasizing the latter.

(9) The ecological effects of organisms, considered as extended phenotypes, are often heritable.

Research in the past three years seems only to have bolstered these conclusions, but I can now see some important nuances.

Last week, I was at Monte Verita in Ascona, southern Switzerland, for a meeting titled The Genetics and Genomics of Eco-Evolutionary Dynamics, organized by a number of postdocs at the Adaptation to a Changing Environment (ACE) centre . The meeting brought together people studying the genomics of adaptation and people studying eco-evolutionary dynamics (two largely non-overlapping groups) to see if some progress could be made toward integrating the two areas of research.

The last light of day from my room at Monte Verita, Ascona, Switzerland.
For some time, it wasn’t clear that such integration was possible or profitable. In particular, because all eco-evolutionary dynamics are driven by phenotypes, the genomics of eco-evolutionary dynamics should simply be the genomics of phenotypic traits – a point that I argued in my 2013 paper (as noted above). However, a problem arises in the case of eco-evolutionary dynamics because the correlation between genes (i.e., genetic variation and evolutionary change) and ecological function is expected to be product of two correlations: that between genes and traits and that between traits and ecological function. Given that correlations are between 0 and 1, this product should be weaker than either of the two correlations. We already know from many studies that each of these two correlations is relative weak, probably nearly always less than 0.5, and so the correlation between genes and ecological function should be VERY weak. In short, the initial perspective of the group was that the genetics and genomics of eco-evolutionary dynamics would be considerably more difficult to study than the genetics and genomics of phenotypic traits (which is already quite hard).

This prospect made most of the genomics people in the room rather less than excited as it seemed to suggest that genomic mapping of ecological function – and the search for candidate genes and causal variants – would be hopeless. This same skepticism was – to some extent – the point I made in my 2013 paper where I argued that the genomics of eco-evolutionary dynamics would be very polygenic and so better served by quantitative genetics. However, further discussion brought up several important points that suggest the tools of genomics might be profitably turned to the exploration of ecological function.

These points can be illustrated by reference to a path model, where a set of genes influence a set of traits which influence a set of ecological functions. In these models, the correlations along each casual pathway are multiplied to get the final correlation between the start (a gene) and end (an ecological function) of that pathway as noted above. However, when multiple pathways link genes and ecological functions, those final correlations are summed across the pathways to get the total effect. Thinking in this manner yields several insights:

The Bailey et al. (2009) version of the relevant path model, which resulted in part from discussion we had when I edited their paper for a PTRSB special issue on Eco-Evolutionary Dynamics.

1. The effects of a given gene on a given ecological function could be greater than the effects of that gene on any one phenotypic trait. This situation could arise when one gene influences multiple traits that each influence the same ecological function. It could also arise if a given gene influences both a trait with a key ecological function and also organismal fitness, with fitness then also influencing ecological function. And the situation is even more promising if more than one aspect of ecological function is influenced by traits and (most obviously) by organismal fitness, such that the total ecological effect could be considerably greater than any single ecological effect.

2. The total effect of all genes on a given ecological function (or on total ecological function) could be large if multiple genes influence one trait that has a strong effect on ecological function(s) or if multiple genes influence multiple traits that together to have large effects on ecological function(s). Further, if we consider a multi-species context, a given ecological function could be influenced by genetic variation in multiple species – and so the total genetic effects of all species on a given ecological function could be large. This last possibility suggests the potential value of analyzing GxG effects on ecological function – as has already been done in several studies where the Gs are different clones. Of particular interest, GxG interactions suggest that the effects of genes in a given species might be most easily revealed if they are assessed on several genomic backgrounds of the other interacting species.  

Seth Rudman gives a summary of the working group. If you look closely, you can see a scribbled version of the path model on the board at left.
If I had the chance, I would now modify the statements in my 2013 paper to make the above points. I would then reiterate that we can treat the ecological effects of individuals as extended phenotypes, and so attempt all of the same genomic work done for more traditional traits (and sometimes fitness): QTL mapping, genome scans (comparing groups of individuals with different ecological functions), genome-wide association studies, candidate gene discovery, and searches causal variants (depending on the specific points of interest). Of course, these methods will need to be combined with quantitative genetic analyses, given that much ecological function will surely be polygenic.

Sadly, I can’t modify the 2013 paper. Nor can I modify the corresponding book chapter given that I will receive the proofs tomorrow and the publisher won’t want me making large changes. However, you can stay tuned for the paper resulting from the discussions in Switzerland, which is being led by Seth Rudman. Go Seth go!


The Genomics of Eco-Evolutionary Dynamics group.

A few days after writing this post, Bob Paine, the originator of the keystone species concept passed away. I took a class from Bob Paine when I was a graduate student and have certainly referred to keystone species multiple times in my writing. In particular, I have argued that eco-evolutionary dynamics are most likely when the evolving focal species is a keystone species (or a foundation species or a ecosystem engineer and so on). Thus, if we are to search for particular genes of large effect, we would probably want to look in these species. And we might then call these large ecological effect genes "keystone genes."

Tuesday, May 17, 2016

Conflict of Interest: Hilborn vs. Greenpeace and beyond


Conflict of Interest is a critical concept in science. The reason is that science is supposed to be objective and not overtly influenced by outside groups or personal interests – at least not beyond the choice of WHAT to study, which is quite reasonably influenced by both. That is, the type of research done is acceptably subject to personal and outside influence but the conclusions drawn from the research should be independent of both. Thus, whenever funding is provided by a particular interest group, one must disclose that funding source.


I am motivated to discuss this topic owing to the recent dust-up between Greenpeace and Ray Hilborn, a professor in the University of Washington’s School of Aquatic and Fishery Sciences. Hilborn is a renowned fisheries scientist who has been very influential at all levels across a broad swath of the globe. A recurring theme in Hilborn’s writing is that frequent claims about over-fishing being pervasive and dramatic are exaggerated. He feels – and his research has often shown – that many fisheries around the world are sustainable, with fish populations often even growing, partly because of sound fisheries management. This picture of fisheries is considerably rosier than the one often painted by some other fisheries scientists by the press and by a variety of environmental groups. Moreover, Hilborn does not pull his punches (the expression “he doesn’t suffer fools” comes to mind) when criticizing the doom-and-gloom perspective espoused by many.

Sockeye salmon caught in the Bristol Bay, AK, salmon test fishery.

Greenpeace feels that overfishing is rampant and that the science supports that conclusion, leading them be at odds with much of Hilborn’s writing. Recently, Greenpeace used the Freedom of Information Act to obtain, from the University of Washington, records of all of Hilborn’s research funding. In a recent online posting, they took Hilborn to task for receiving funding from the fisheries industry but for not always reporting these contributions in his writings about fisheries. They provided some specific examples of papers published in a range of journals – and of articles in the popular press – where Hilborn came to the conclusion (or expressed the opinion) that overfishing concerns were exaggerated; but in which funding from the fisheries industry was not acknowledged. The implication of their article was that Hilborn was somehow influenced by the industry to paint a rosier picture of fishing than was actually the case.

Hilborn defended his record in several ways. First, he provided a pie chart of all of his funding over the period in question, which shows a diversity of sources, including the fishing industry (13% of the total). He suggested it was unclear what sort of bias he was supposed to have given that his research has been supported by government, environmental NGOs, community groups, foundations, the fishing industry, and others. Second, he noted that the accepted practice in science is to report funding sources for the specific work being reported in a given paper, not all sources of funding one has ever received. The latter scenario would obviously be unrealistic as the list of sponsors would be longer than some of the papers. Third, he pointed out that the actual data are not really be questioned, nor are his research methods and approaches.


I took a personal interest in this story because I did my PhD and MSc at the University of Washington in their School of Aquatic and Fishery Sciences (then the School of Fisheries). I worked in Alaska for the school’s Alaska Salmon Program, which received some funding from the fisheries industry. More directly, I worked for several summers at a remote field camp in Alaska with Ray Hilborn and his family. Finally, I published two papers with Hilborn based on work done by students with who we both collaborated (Gende et al. 2004 – Oikos; Carlson et al. 2007 – PLoS ONE). Throughout all of this, my experiences with Ray have always been positive, I have the utmost respect for his science, and I have never had any thought that he might be influenced by industry when conduct his science or expressing his opinion. In short, like many other scientists, I am fully in support of Hilborn and his science.


Yet the broader issue isn’t so cut-and-dry. A critical point about conflict of interest is not just that you shouldn't have one (or that you should state it clearly if you do), but also that you need to avoid the PERCEPTION that you might have a conflict of interest. Indeed, some journals (such as Science) ask you to disclose any funding sources that COULD BE PERCEIVED as a conflict of interest. And here I think that Hilborn – and everyone else – can probably do better.

I worked for 10 years - some of them with Ray Hilborn - at this University of Washington camp on Lake Nerka in Alaska. It's development, maintenance, and operation was partly funded by the  Pacific Seafood Processors Association.
Consider the generally accepted practice of only disclosing funding sources in a paper that are specific to the work being reported in that paper. Now imagine an extreme situation where a particular industry (let’s say tobacco) funds nine out of ten of your papers. The tenth paper is also relevant to that industry (it shows that a particular chemical found in tobacco does not have negative health effects) but it was based on research not directly funded by that industry (perhaps NIH funded it.)  I suspect that a number of people would quite reasonably worry that your tenth paper was influenced by the tobacco industry even if it wasn’t directly funded by it. Thus, I suppose an outsider who doesn’t know Hilborn and his work might feel that not disclosing, in every paper he writes about fishing, that he has received considerable funding from the fishing industry is misleading. Similarly, others might think it misleading if he doesn’t disclose his considerable funding from environmental groups or from government.

Of course, every paper one writes can’t list every specific funding source that one has ever received – it would indeed be too cumbersome. However, it would certainly be easy (and short) to simply list the broad classes of funding sources that one has received, perhaps even overall amounts for each of those sources. Indeed, Hilborn’s pie chart makes clear that he isn’t “in the pocket” of industry or anyone else for that matter – and Hilborn and his colleagues have been using this to great effect. Combine broad-brush historical reporting like this with a listing of the specific funding sources for the research in the specific paper, and you should be covered.

The main reason the PSPA funded the UW work in Alaska was because we ran a test fishery in Bristol Bay to help predict salmon returns. I helped with this test fishery in two years. Just after the second year, the ship sank and everyone onboard died, including the gentleman in this picture, Blake Grimstein. He was our able, helpful, and congenial Captain for the test fishery.


In closing, I doubt very much that Ray Hilborn is influenced by the fishing industry in the conclusions he draws about fishing. I also think that his research is likely to be the most defensible of any out there. At the same time, I can see how outsiders might PERCEIVE a conflict of interest, which might cause some people to think a bias exists. Thus, I suggest that in any situation where people might worry about a conflict of interest in your work, that a sentence be added to the end of your acknowledgements that simply says something like: “Our research programs have been funded by a diversity of sources, including governments, foundations, NGOs, community groups, and industry.”

Notes:

1. Both of the papers I published with Ray Hilborn acknowledged industry funding. However, none of those papers had anything to do with fishing, and so I doubt anyone would care regardless.


2. I have a long personal and professional relationship with Hilborn, as detailed above. We have also repeatedly talked about working on grants for future work in Alaska. And - perhaps most importantly - I follow his daughter, Anne, on twitter. 

3. I have, occasionally, been a dues-paying member of Greenpeace, which I think has done some great (and some not-so-great) work in raising the public’s consciousness about environmental issues.

4. Conflict of Interest is also important to discuss in other aspects of science, such as reviewing and editing - but I do not touch on those aspects here.

Friday, May 6, 2016

How to be a (new) professor

I have been marching through a long series of “How to” posts for young scientists. The last few posts have been about getting a faculty position, so it now seems appropriate to ask what you should do when you get one. I suppose the post could have been titled “How to get tenure,” but the title I chose emphasizes a process rather than an end point – and many aspects of the process continue following tenure. Nevertheless, I will follow this “How to be a (new) professor” post with a later one on “How to be an (old) professor”, with the distinction roughly being pre-tenure versus post-tenure.
Please bear in mind that I am not saying I do all of these things well – in fact, I did/do some poorly, as I am sure my colleagues and students can attest.



Try to postpone your start date. Nearly every faculty member I have spoken to at my or any other institution has complained that their lab wasn’t ready when they started their position. Indeed, no matter what time frame the contractor/university gives you, it will take longer – sometimes much longer. This problem will delay the start of your research program – sometimes substantially - and can be very stressful for new faculty members. One way to reduce this problem is to postpone your formal arrival at the university as long as you can. The typical postponement is 1-1.5 years depending on when you accept the position, but I have seen some that are even longer. This strategy increases the chance your lab will be ready, allows you to do more postdoctoral work (the benefits of which are extolled here), allows you to recruit students, and gives you an opportunity to submit research grants. All of these outcomes help you to hit the ground running when you arrive. Of course, delays can persist even if you postpone your arrival – but a postponement greatly helps. Also, note that some universities or departments might not allow postponing your arrival.

Take a teaching/service holiday if you can get one. All research universities want you to get a good start on your research and so will grant you a one-year postponement of teaching and service responsibilities. You should definitely take this as concurrently building your research program and teaching/developing new courses is a recipe for stress and for subpar performance in each area. However, note that postponing your arrival (the above suggestions) often negates additional postponement of teaching and service.

Source - but see the end of this post

Don’t be too greedy. Advice about negotiating faculty positions often emphasizes how you should bargain hard and get as much as you can possibly get out of the department and university. I don’t agree. Bargaining too hard makes you seem selfish, arrogant, and greedy, and can thereby damage relationships with your Dean, your Chair, and your colleagues. As one direct example, you might hold out for more space and you might get it; but, unless a lot of free space exists in the department, getting more space can mean taking space away from your colleagues – and they will know it. Also, a larger start-up can mean the Department has to take money from some other activity. The same general point can apply to salaries. If you negotiate a huge salary, and your salary is known (as is often the case), faculty who have been around longer but are paid less can resent it. I suggest asking for similar amounts of space as other faculty in the department with comparable research and for similar salaries to other recent professors.


Say yes to all requests that help your colleagues and don’t hurt you. As soon as you are a faculty member, you will immediately be besieged by requests to be on various student committees. I would say yes to ALL of these. Here you have a great opportunity to help your colleagues and their students at minimal cost to yourself. I say the cost is minimal because the time you invest in being on a student’s committee is trivial in the big picture (I would guess 2-4 hours – to read a proposal and sit in the meeting – per student per year). Moreover, turning down a request to be on a committee can offend the faculty member who suggested to the student that you be on the committee – and that is never a good idea.

Take some students right away. Sometimes the temptation is to hold out for the “right” student, but being too picky can excessively delay your research program and can act against you in reappointments and grant applications. Moreover, determining the “right” student is hard, if not impossible. I personally think that success in graduate school is hard to judge at the outset, no matter how carefully you attempt to vet a given student. Indeed, I have heard many instances of faculty who were absolutely sure a student would be great, only to have the student fall far short of expectations. Conversely, I have heard many other instances when a student who was unimpressive in interviews turned out to be outstanding. Thus, it is much better to take a student or two in your first year than to try to hold off for who you think will be the perfect student.

Don’t be too demanding of your first students. New faculty members have relative few students, and so the early development of their research program depends heavily on those students. Moreover, new faculty members have relatively few other demands on their time. As a result, new faculty often interact a ton with their first students. This intensity of interaction can be great for the student, but it can also be problematic if the faculty member relies to overtly on the student for the success of their research program. This kind of pressure will not help your students – and therefore won’t help you either. Try not to micromanage. Try not to tell them that your next grant or tenure or whatever depends on their success. Related to this point, new professors tend to have over-high expectations for students, partly because they presume their students will be like themselves. Remember, however, that the average professor trains only a few future professors and so, in principle, only a very few of your students will be as successful as you were. Regardless, all of your students will have unique skills and contributions – and all can form a valuable – indeed vital – part of your research program.

Don’t spend your start-up money too quickly. I screwed up on this one. My view as a new professor was that I should immediately have my lab fully set-up to match my vision of what my lab would need for the next ten or more years. I therefore bought really expensive (top-of-the-line) equipment based on my expectations of what my research program would look like – before I actually had students doing research! As a result, some of the equipment turned out to be not of much use or was overkill because my research program morphed rather quickly depending on funded grants and student interests and so on. If I had hoarded my start-up and only bought what I needed when I needed it, I would have had considerably more flexibility for longer. (However, make sure you know the deadlines and requirements for spending your start-up – or it can be taken away from you.)

Focus on publishing and research. Evaluations of new faculty, including for tenure, are usually based on some weighting of research, teaching, and service. At a research university, the first of these categories is usually formally weighted the most and informally weighted by far the most. You must be a good researcher, the primary measure of which is publications (and often grants). Hence, most of your time should be dedicated to research and its publication. Teaching is also important but, to be honest, most reappointment and promotion decisions merely require you to be an adequate teacher teaching an adequate amount. I suggest you teach the required amount (but not more) and that you concentrate on quality over quantity in your teaching. Importantly, however, the TIME investment should be greatest in research – and within that arena it should  be greatest in publications and then grants. As for service … well, do your share but don’t seek this out – it will come to you soon enough.


Participate in community-building social activities. The extent to which people like to engage in social activities at work is highly variable and some people can resent (or at least be very uncomfortable with) continual encouragement to participate in those activities. If social activities make you really uncomfortable, then you certainly shouldn’t overdue them. However, social activities that build a sense of community in your unit are an extremely important way of integrating into a department, building collaborations, sharing ideas, and helping to make your department more collegial and more than just a collection of individuals.

Seek out interactions and collaborations with your colleagues. One of the most rewarding aspects of being in a particular university department is interacting and collaborating with your colleagues – I love it. Try to meet with and discuss all aspects of research and academic life with your colleagues. Seek out and pursue logical collaborations and build the familiarity networks that might allow future collaborations. Of course, some of your colleagues may come to annoy you (not my colleagues, of course) and some of the collaborations might not work out; but those failures should be more than made up for by the successes.

The network of interactions among ecology and evolution faculty at McGill circa 2007 (courtesy Gregor Fussmann). Each circle is a person and each line represents at least one paper or grant. Many more links would exist now.
Have lab meetings, perhaps jointly with other professors. Lab meetings are an essential part of building a community within your lab and sharing knowledge and ideas – many rewarding collaborations and discussions and arguments and opportunities and papers have come from my lab meetings. However, a new faculty member often has very few people in the lab, which reduces some of these benefits. One option is to join the lab meeting of another professor with a similar or complementary set of research interests. These joint lab meetings can really help your students and will help build bridges across laboratories.

Social events with your students. Following from the above two points, a sense of community in your lab will be enhanced by lab social activities, such as going out for drinks or lunches or having Christmas parties or lab retreats. These activities not only foster friendships with your lab but they make your students more comfortable interacting with you – because they now know you a bit outside of work. However, it is critically important to not engage in, or promote, any activities or discussions that could be considered (or perceived as) harassment in any form. So make sure that the social events are in appropriate settings and at all times carried out with respect and equality.  

Relax - it will be OK. For some reason, new faculty worry tremendously about tenure. However, tenure rates are extremely high and so the chances are very good that you won't have any problem. You don't need to kill yourself. Although perceptions are that faculty work extremely long hours, which can be the case, many professors have a nice work-life balance with a reasonable number of working hours. Being a faculty member is probably something you have long dreamed of - so make it fun and rewarding, not stressful and demanding!

I could write more but this post is already rather long, so I will save more ideas for an upcoming post on “How to be an (old) professor.”

Here is a link to the earlier "How to" posts, some of which are shown below.