Evolution 101: Group Selection

Posted on January 20, 2014 by Danielle Whittaker

This week’s Evolution 101 post is by MSU graduate student Thomas LaBar.

A wolf pack hunting as a group. (Source: Doug Smith; Wikipedia)

A wolf pack hunting as a group. (Source: Doug Smith; Wikipedia)

Many organisms live and interact within groups. Beehives, wolf packs, and bird flocks are just a few of the groups that exist in the natural world. Humans also live in groups of many different sizes. However, even though groups are common in nature, early researchers of evolution struggled to explain how they evolved.

The problem was that individuals within groups often cooperate and act altruistically, or nice, towards each other. The idea of natural selection, where individuals compete and the fittest ones get to pass on their genes, didn’t appear to allow for cooperation. Group selection, the idea that groups compete just as individuals do, and the behavior of the individuals within a group affects each individual’s own fitness, arose to explain the evolution of altruism.

beehive

A queen bee in a beehive. (Source: Waugsberg; Wikipedia)

Biologists define an altruistic behavior as one an individual takes that lowers its own fitness and increases the fitness of another individual. For example, consider how reproduction works in a honeybee colony. Only one female reproduces: the queen bee. The rest of the females, the workers, help raise her offspring instead of producing their own. If evolution occurs by individuals passing on their genes, how did this behavior evolve? This type of behavior isn’t unique to honeybees; many species possess altruistic behaviors. However, if an altruistic behavior lowers an individual’s fitness, and evolution selects for individuals with higher fitness than others, how does altruism evolve?

The original explanation group selection provided was that individuals would cooperate for the “good of the group,” or the “good of the species.” In this scenario, group of altruistic individuals has a higher fitness, due to altruism, than a solely non-altruistic group. These altruistic groups would outcompete the non-altruistic groups, just like an individual with higher fitness (in an evolutionary sense) outcompetes individuals with lower fitness. However, this argument has a flaw. Imagine that one of the individuals in the altruistic group has an offspring with a mutation that causes it to be non-altruistic. This individual would have a higher fitness than the other group members, and, over time, future generations of this group would consist of an increasing number of non-altruistic individuals. Eventually, the group would become solely non-altruistic.

This issue led to the demise of group selection for many years. In its place arose kin selection theory. This theory explained the evolution of altruism by showing that it could evolve when the relatedness of individuals to each other was considered. If individuals acted altruistically towards relatives, the altruistic individuals’ genes would still spread, since its relative shared many of the same genes, compared to other non-relative individuals. 

After kin selection was discovered, very few biologists proposed group selection ideas in their research. However, the theory of group selection has seen a revival in recent years, often under the term multi-level selection theory. As the name implies, this theory argues that one must consider not only the fitness effects of the individual, but also the group. One reason for group selection’s revival is that, in some cases, it is hard to tell which theory (group or kin) provides better explanations. Again, consider the beehive. All of the worker bees are related to the queen. Is it better to use group selection or kin selection? Furthermore, in some cases, the mathematics of the respective theories (which is beyond the scope of this post) allows researchers to draw the same conclusion. As it stands today, both theories have scientists that support them. One should try to get a balanced view when studying this topic, as both sides have interesting ideas.

Further reading: 

Posted in Evolution 101 | Tagged , , , , | Leave a comment

Creating Life and Worlds: Solving Evolutionary Problems in the Digital World

Posted on January 15, 2014 by Danielle Whittaker

This blog post was written by students in BEACON’s Fall 2013 Computational Science for Evolutionary Biologists course, taught by MSU faculty Titus Brown and graduate student Randy Olson. Blog post lead author: S. Kevin McCormick, with contributions from Zach Laubach, Nicolas Schmelling, Josephine Wee, and Phillip Brooks.

 

Ever wanted to play “god”? Perhaps, create a world and place some simple organisms on it to see what develops. You could surely answer a lot of questions about the beauty and diversity that has evolved on this planet. Some of our own have actually managed to run long term experiments on fast replicating organisms to do just that. However, that’s not exactly possible when one is working with longer living animals, or most other living organisms for that matter.

Another option would be to create artificial life forms that live and evolve within a digitally created world. This is not science fiction: computer models allowing scientists to test how populations of organisms react to change started popping up decades ago. Now labs are creating digital organisms with “DNA code” that can replicate, mutate, and evolve under any scenario that computer scientists and biologists can cook up. During Michigan State University’s CSE 801 course, run through the BEACON program, graduate students are introduced to a few of these computer models. One is a digital optimal foraging model called Area Restricted Search (ARS). The ARS model is based on a genetic algorithm that allows researchers to create a population of organisms with a few simple traits that can mutate over successive generations. These populations of organisms are placed in a world with a resource type, and given an award of higher fitness if they can find the resources. Natural selection can then act upon the population by giving preferential replication to the organisms with higher fitness.

Our class was first taught how to create digital populations that both did and did not evolve ARS strategies to find resources. We accomplished these tasks by allowing for trait changes in turning angle when in contact or away from resources, and the memory to turn in those situations, after each successive generation. Typically, when resources were sparsely and randomly distributed, an ARS strategy evolved quickly. Organisms turned more when in contact with food, turned less off food, and memory capacity increased. However, when resources were ubiquitous through the environment, ARS did not evolve, or evolved very slowly. For our class’s final project, we were asked to take the code for the ARS model, and manipulate it to answer evolutionary questions we were interested in. Our group wanted to see if we could evolve different ARS strategies in populations that were allopatrically isolated in environments with different resource types. If this was possible, we then wished to determine if the different ARS strategies of these populations gave advantages to a particular population when introduced to an environment with different resources. We hoped to use this as a model for testing questions about indirect competition.

Figure 1: Example of environmental switching. Blue = small valuable resources, Green = large low value resources.

Figure 1: Example of environmental switching. Blue = small valuable resources, Green = large low value resources.

We began by manipulating the IPython code to create different resource environments that could change at a time our choosing. The environments themselves are made of digital grids, called ipythonblocks, where resources are distributed randomly over each successive generation. We altered the code to allow for multiple resource types of different size, value, and presence in the environment. We then created four simulations to test if allopatrically isolated populations evolved different ARS strategies, which could give one population an advantage when introduced to its “competitors” environment, or to environment of mixed resources (Figure 1). Each population (30 individuals) was first allowed to evolve for 50 generations on either rare small high-value resources, or large common low-value resources, at a trait mutation rate of 0.01 per generation. Then these populations were switched onto a new resource environment and allowed to evolve for another 50 generations under the same mutation rate. These simulations were run through EC2 instances on Amazon mainframes, and each took only 8 hours to complete.

Unfortunately, this first experiment did not work as planned. Separate ARS strategies and traits never evolved to fixation, and limitations in our code led to treatment effects in our separate environments (Figures 2 & 3). Most notably we ended up doubling resources when we created mixed resource environments, and we may not have allowed for enough time for separate ARS strategies to evolve. Therefore, because we were allowed another week, we chose to extend our model and fiddle with our created worlds and organisms a little more.

Figure 2: Fitness level over 100 generations. Shift at 50 generations reflects switching each population’s environment. Lines represent each individual population mean. Bars represent 95% confidence intervals.

Figure 2: Fitness level over 100 generations. Shift at 50 generations reflects switching each population’s environment. Lines represent each individual population mean. Bars represent 95% confidence intervals. 

Figure 3: Turn angle increase when encountering food over 100 generation. Lines represent each individual population mean. Bars represent 95% confidence intervals.

Figure 3: Turn angle increase when encountering food over 100 generation. Lines represent each individual population mean. Bars represent 95% confidence intervals.

First we created worlds where opportunities to increase fitness were equal in all aspects except for the ability to find the resources. Then we ran simulations we previously did not have time for, including evolving populations on mixed resources and switching them onto monocultures, as well as evolving populations solely on one type of resource base as controls. Next we allowed populations to persist for additional generations before switching resource environments.

Looking back now, I think that this may have been enough, but it seemed that the power to create and modify our little worlds with a few more lines of code may have gone to our heads. We chose to increase the mutation rate 10 fold prior to the environment shift, thinking that it would allow for a greater chance of different ARS strategies to evolve in the time we had. We then went even farther and stopped mutation just prior to the environment shifts. It was our hope that, if different ARS strategies did evolve to fixation, populations would be forced to use their evolved strategy in their new environment without the possibility of evolving a new one. Ideally this would artificially reflect wha

t is occasionally seen when generalist foragers invade new environments, and when specialist populations are displaced into new environments.

Unfortunately that last code change to the mutation rate was just a bit too much. By increasing the mutation rate tenfold we allowed for trait changes to jump all over the place for each successive generation. As Randy said, “we more or less eliminated inheritance,” as each individual lineage could drop or increase in trait changes at a rate higher than the code could allow for fitness increases. Therefore, there was no way for natural selection to properly act upon the population, perhaps some individual lineages, but not the population. This can be seen in Figure 4, when we shut off mutation at 60 generations, because afterward we observed what appeared to be population sorting. Following the mutation shutoff, individuals within the population with “better” trait values started to out-compete their relatives.

Figure 4: Fitness level over 100 generations. Lines represent each individual population mean. Bars represent 95% confidence intervals. Note: the blue line drop just before 70 generation reflects a coding error

Figure 4: Fitness level over 100 generations. Lines represent each individual population mean. Bars represent 95% confidence intervals. Note: the blue line drop just before 70 generations reflects a coding error.

It seemed that everything we did to try to create different ARS strategies had the opposite effect. All traits, averaged across each population, ended up at the same level prior to the mutation shutoff, and the only significant difference we observed was different levels of fitness for each resource environment. It seems that playing “god” is a bit more difficult than we thought. However, thanks to this course we now have a better understanding of what it takes to create digital organisms and worlds, as well as how complicated it can be to model evolution. In the future we hope to extend these models to allow for other drivers of evolution, such as direct competition, death, energy expenditure, stochastic resource availability, evolution of different traits…….    

Posted in BEACON Researchers at Work | Tagged , , , | Leave a comment

Evolution 101: Digital Evolution

Posted on January 13, 2014 by Danielle Whittaker

This Evolution 101 post is by MSU graduate student Armand Burks.

When scientists study the process of evolution in living organisms, two of the key limiting factors are that of time and the amount of data available. In practice, it is necessary to investigate over many years of the genetic history of the organism of interest in order to gain meaningful insight into the many phenomena related to its evolution. In addition to being able to look back far enough, scientists need to have as much genetic data as possible, from many organisms and their ancestors, in order to understand their relationships and how evolution has affected them over time.

Technological advances such as genome sequencing and many others have definitely made it possible to study evolution at a much greater scale than was possible before. However, it is quite often the case that there is simply not enough genetic data collected, or that the available data doesn’t span back far enough to answer all of the questions related to the evolution of the organism of interest. To sum it all up, evolution typically takes a lot of time. In an attempt to alleviate some of these problems, Digital Evolution is an approach that allows researchers and enthusiasts to study various phenomena related to evolution in a simulated computerized environment.

Digital Evolution in Research

Avida is a digital evolution software tool that was pioneered by Charles Ofria, Titus Brown, and Chris Adami. In Avida, digital organisms live in a virtual environment where they replicate, mutate, and compete with each other for survival and resources. With this very simple concept, researchers are able to conduct experiments to study phenomena that occurs in real biological systems. Although it is only a biological simulation, researchers can very quickly test hypotheses and get results that, if possible in nature, would have taken many years. Thus, digital evolution allows researchers to conduct experiments and obtain results in their own lifetimes, which directly adds to the body of knowledge on evolution at a pace that is not possible by observing nature alone.

Screenshot of an Avida run

Screenshot of an Avida run

 

Digital Evolution in Teaching

Not only does digital evolution allow scientists to study evolution at an unprecedented scale and speed, but it also serves as a great tool for teaching the concepts of biological evolution to students. Avida-ED is another digital evolution software tool that comes in a much simpler form than Avida. Avida-ED is an educational tool that aims to be more approachable by a much wider audience. This software can be used to teach students about the concepts of evolution, providing hands-on classroom experience, and allowing students to conduct their own small experiments as they learn.

A screenshot of the Avida-ED software. Here, we can see a population of digital organisms in the black square, and various tools in the user interface.

A screenshot of the Avida-ED software. Here, we can see a population of digital organisms in the black square, and various tools in the user interface.

As we can see, digital evolution is a very useful tool in both understanding and researching evolution. Digital evolution software allows researchers to study evolution at a much larger scale, and much more rapidly than is possible in nature. By taking advantage of the many technological advances that computers provide, digital evolution allows researchers to study the various phenomena of evolution while overcoming the issue of time. It also provides an intuitive and approachable means for teaching and learning about evolution at various levels. Although this article only describes two digital evolution tools, there are several other tools with varying levels of complexity and generality.

Posted in Evolution 101 | Tagged , , , | Leave a comment

Evolution 101: Beauty is in the Genes of the Beholder

Posted on January 6, 2014 by Danielle Whittaker

This Evolution 101 post is by MSU graduate student Anselmo Pontes.

Peacock by Jebulon /Wikimedia Creative Commons License.

Peacock by Jebulon /Wikimedia Creative Commons License.

What do you think when you see a woman painfully balancing on sky-high heels? How about when you overhear the lame pick up lines guys come up with to get a girl’s attention? We do all kinds of things to impress the opposite sex. Though we can pass the limits of good sense at times, we can’t help but try. What we may not be aware of is that the way we alter our behavior and appearance can change the course of evolution.

Take the guy with the pick up lines. A woman may find him witty, an indirect sign of intelligence, and possibly a resourceful mate. If he manages to partner with the woman and have children, his characteristics will live on. More subtly, the woman’s preference will also be passed on to the next generation. So, this couple will tend to have boys that come up with silly pick up lines and girls that go for them.

The first one to notice this relationship was Charles Darwin. He observed that the choices that animals made when mating affected their descendants. He considered this to be a force of evolution akin to natural selection, and called it sexual selection.

Darwin cited the peacock’s huge, colorful tails, which serve no other purpose than to woo females and may even make them vulnerable to predators. However, peahens, with their discreet plumage, are the ones selecting the mates. Since they prefer males with large and bright tails, these are the kind you would expect to see in the next generation. Over time, the tails tend to become larger and more colorful (and female’s preference for large, colorful tails increases in tandem).

This happens because some preferences, physical characteristics and behaviors are heritable traits, coded in the genes. Therefore, to a choosy female, these big, beautiful tails indicate either good genes and health or a desirable trait that will give her offspring an advantage in future mating efforts.

In most species, the female does the choosing. This is because females are usually the ones that make the most investment in the offspring. Males are a dime a dozen and rarely stay for child rearing. Females are the ones that tend to and nurture the young.

The male bowerbird builds elaborate structures to attract females. Photo from mudbayworldwonders.blogspot.com.

The male bowerbird builds elaborate structures to
attract females. Photo from mudbayworldwonders.blogspot.com.

It isn’t only physical beauty that gets attention. Some species attract mates by offering gifts of food, building elaborate nests, singing and even dancing. Bowerbirds, a native of Australia and New Guinea, create masterfully ornate structures akin to theatrical stages called “bowers” with the single function of impressing females. A male bowerbird spends days, sometimes weeks collecting twigs, flowers, rocks, seashells and assorted colorful objects and assembling them into a carefully decorated structure to demonstrate their individual prowess and good taste. Females visit multiple bowers and return to the one they judged the best.

Sexual selection is not only about attracting mates, it’s also about fighting for them. In many species, males fight with each other and only the winners get to mate – sometimes with many females.  Such is the case of sea lions and deer, where winners fight to maintain a harem of females, and losers don’t get to mate at all.

The elk’s large antlers intimidate other males and ensure a harem of females. Photo from ZaPrirodata /Wikimedia Creative Commons License.

The elk’s large antlers intimidate other males and ensure a harem of females. Photo from ZaPrirodata /Wikimedia Creative Commons License.

Since big antlers intimidate lesser opponents, and long colorful tails attract mates, these traits tend to continue developing in a sort of “arms race” called runaway selection. Eventually though, antlers will grow too heavy to carry and the drag on a long tail will outweigh its advantage. This is when sexual selection tapers off, and traits favoring survival become more important.

Humans have also been shaped by sexual selection. It is suspected to have influenced the size of our brain, the strategic placement of body and facial hair as well as our height and tone of voice. So back in the cave, when the choice was between the rude, scruffy-looking ape and that caring, sleek someone, our ancestors’ pick may have determined how we turned out to be the smooth, quick-witted, fabulous-looking creatures we are.

Posted in Evolution 101 | Tagged , , , | Leave a comment

BEACON Researchers at Work: Extreme science in Siberia

Posted on December 17, 2013 by Danielle Whittaker

This week’s BEACON Researchers at Work blog post is by MSU graduate student Paul Wilburn.

Paul WilburnA cold shiver and briefly clenched teeth are common quick reactions displayed by new acquaintances when informed that I drill lake ice in Siberia for a living. Perhaps mental images of cold, biting winds, clumsy mittens and tools that freeze from exposure can instantly overwhelm. They are likely the opposite of a dream job, like collecting tropical water samples to bring back to a cooler under a palm tree on Carefree beach. But then I turn their attention to memories of Jack London’s White Fang or Call of the Wild. Recall the vast Nordic landscapes, endless frozen boreal, divided into wolf pack territories by the Yukon, Klondike and the McKenzie rivers. Where wolves, dogs and humans alike are forced by the unforgiving wild to find their place in a primal land, bound by the law of meat. It is a place where distractions are unthinkable, where survival, competition and procreation come to the surface. Life and death are exposed, and the perseverance to exist is more straightforward. The explicit nature of this system is what attracts biologists, like me, to uncover what it takes to survive the harsh abiotic conditions for some of world’s most exotic organisms. To answer this question, I study extremophile microorganisms endemic to Lake Baikal, Siberia.

Baikal seal

Baikal seal

Lake Baikal is a UNESCO world heritage site and the planet’s oldest, deepest and most voluminous lake that holds about 20% of the world’s unfrozen freshwater. It is historically cold year-round, although the region is predicted to experience rapid warming. Baikal has about the surface area of, and is ten times deeper than, Lake Michigan. And while the American Great Lakes were formed by melting glaciers about 15-10,000 years ago, Baikal fills a tectonic rift and clocks in at 30 million years old. Partly because of its age, Baikal is an island of biodiversity, marked by high endemism from algae to the top predator, the world’s only freshwater seal that lives only in lake Baikal and nowhere else. Endemic organisms, like the seal, are Baikal specialists that evolved over millions of years to dominate the ecosystem at all trophic levels. How do they do it? Given the millions of years of evolution, endemic species likely developed streamlined metabolic adaptations to unique Baikal conditions, like constant cold. Indeed, Baikal is frozen for four to five months out of the year, and the peak summer temperature goes above 10 degrees C for perhaps a week or so. Thus, I am particularly interested in molecular adaptations of Baikal endemic microorganisms to constant cold.

Research cruise

Research cruise

The first part of my thesis project is to determine the distribution of endemic microorganisms at multiple locations in the lake. While the open waters are dominated by endemics, the shallow bays, some of them quite large, along Baikal’s 2000-kilometer coastline are inhabited by microorganisms with cosmopolitan distribution. But no one has used molecular tools to systematically survey the lake and determine relative abundances of endemic and cosmopolitan taxa at different locations. In Summer 2013, I joined a 12-day “circum-Baikal” research cruise and collected samples from 52 locations along with abiotic metadata on temperature, light and nutrient content. Using amplicon metagenomic methods, I will estimate relative abundances of endemic and cosmopolitan taxa and associate them with the abiotic conditions, like temperature. I would expect endemic and cosmopolitan organisms to dominate cold and milder locations, respectively. The data will also tell me something about standing variation in various microorganism populations, and thus their potential for evolution in the scenario of rapid disturbance, like warming.

Drilling into ice on Lake Baikal

Drilling into ice on Lake Baikal

The second part of my thesis is to get some insight into the actual adaptations of Baikal endemics to constant cold. In Winter 2013, I spent three weeks on frozen Baikal, drilling the ice and collecting samples for shotgun metagenomic analyses (Figure 3). Shotgun metagenomics is a method to sequence and assemble draft genomes of all organisms in the collected sample. About two dozen molecular adaptations to constant cold have been described for polar organisms, like Antarctic fish and diatom algae, in the literature. These adaptations are coded in the genome and will be visible in the assembled data. Examining draft genomes of endemic organisms will give me clues into possible strategies used by endemics to successfully dominate Baikal.

For the third part of my thesis, I will use transcriptomic methods to refine a mechanistic picture of how endemic and cosmopolitan organisms deal with cold and warm temperatures. For this step, I will grow microorganisms in culture and look at the genes they express. With metagenomic information as reference and transcriptomic information on gene expression, I will be able to tell if endemics use some genes more or less than cosmopolitan species. These findings will experimentally confirm proposed strategies used by contrasting microbial taxa to grow in constantly cold conditions.

Shallow bays and small lakes all around Baikal are full of weedy temperate species. What keeps them from invading and overtaking the pristine Baikal ecosystem? What happens if Baikal warms up in the near future? These questions are central to my thesis. Understanding the molecular mechanisms of cold adaptation in Baikal endemic and cosmopolitan species will give a mechanistic picture of their differences. Future work will focus on quantitatively predicting the relative success of these different microorganism groups in different environments.

For more information about Paul’s work, you can contact him at wilburn4 at msu dot edu.

Posted in BEACON Researchers at Work | Tagged , , , , , , | Leave a comment

BEACON Researchers at Work: Mating System Evolution

Posted on December 9, 2013 by Danielle Whittaker

This week’s BEACON Researchers at Work blog post is by Michigan State University postdoc Sarah Bodbyl.

Sarah Bodbyl“Bees do have a smell, you know, and if they don’t they should, for their feet are dusted with spices from a million flowers.” ― Ray Bradbury, Dandelion Wine

Recent and dramatic declines of pollinating insects have been observed all over the world by scientists and amateurs alike. Human-induced environmental changes, such as habitat loss, climate change, invasive species introductions, pesticide use, and novel disease transmission are all thought to be contributors to this global pollinator decline. Since many (at least 180,000) plant species rely on pollinators to reproduce via seeds and fruit, pollinator declines have folks worried about both wild and agricultural plant populations.

One of my research areas investigates the question, what happens to plants when they lose their pollinators? Specifically, I want to know if plants experiencing rapid loss of their pollinators are able to adapt and maintain viable populations by increasing their ability to reproduce without pollinators.

Mating system evolution

Flowering plants, or angiosperms, have evolved numerous reproductive strategies to overcome their sessile nature. Outcross fertilization (outcrossing) is a sexual reproduction strategy occurring when pollen from the flower of one plant is transferred to the flower of another plant. Self-fertilization (selfing, or autogamy) occurs when pollen from one flower fertilizes the ovaries of the same flower, or another flower within the same plant. Some plants are also capable of asexual reproduction via apomixis or vegetative cloning, but my research focuses on sexual strategies. Mating system is the degree of allocation by a plant population or species to each reproductive strategy. Some plants exclusively outcross, some exclusively self, and others employ both in varying degree; termed mixed mating. Both outcrossing and selfing can occur either with or without the aid of pollinators, but many flowering plants have evolved mutualistic relationships with pollinators to facilitate outcrossing.

Mimulus beeAs in most variable traits found in nature, plant mating systems are not fixed and evolve in response to selective pressures. The evolutionary transition from outcrossing to selfing has occurred repeatedly in angiosperms and is associated with changes in floral biology, life history, and ecology. Some evolutionary biologists have predicted that plant populations that lose their pollinators may be capable of evolving increased selfing capacity, a mechanism of reproductive assurance. However, this transition has never been directly observed taking place in a plant population in response to pollinator loss, so I decided to try to recreate the process in an experimental setting.

Experimental evolution of mating system in monkeyflowers

To determine the immediate effects of pollinator loss on a plant population, my doctoral advisor (Dr. John Kelly, University of Kansas) and I set up a large (~40,000 plant) experimental evolution study, using a primarily outcrossing North American wildflower, Mimulus guttatus (monkeyflower). These cute yellow-flowered plants possess highly variable morphology and life history traits and are primarily pollinated by native bumblebees.

As a starting point for experimental evolution, we used seed from a large source population to create two experimental treatment groups, each containing two replicate populations: “Bee” populations (B1, B2) received abundant bumblebee (Bombus impatiens) pollinators while “No Bee” populations (A1, A2) experienced a complete lack of pollinators. We allowed these four initially equivalent populations to evolve for 5 generations under the two pollination treatments in a greenhouse.

Population fitness over 5 generations of evolution. B1 and B2 are Bee populations, A1 and A2 are No Bee populations. Inset – M. guttatus flower.

Population fitness over 5 generations of evolution. B1 and B2 are Bee populations, A1 and A2 are No Bee populations. Inset – M. guttatus flower.

We recorded plant phenotypes and genotypes in the final generations of the study and were impressed with the speed of the evolution observed within and among the treatments. Populations without pollinators (No Bee) displayed low and declining fitness (measured by seed production) in the early generations of the study but then rebounded, culminating in a ten-fold increase in self-fertilization capacity compared to the initial ancestral population (see figure at right). This pattern—a rapid increase of adapted genotypes after an environmentally induced fitness crash—is consistent with the theory of evolutionary rescue. Thus, at least under our experimental conditions, we demonstrated that rapid mating system evolution via increased selfing capacity is a potential adaptive strategy for plant populations experiencing pollinator loss. Genotyping of our experimental plants at multiple loci revealed rapid allele frequency changes occurring in the No Bee populations, which were correlated with the evolving phenotypes.

Bivariate plot of the negative relationship between mean herkogamy and mean self seed by experimental population after 5 generations. Error bars are +/- 1 SEM. B1 and B2 are Bee populations, A1 and A2 are No Bee populations, and Source is the original population without treatment. Inset—cross-section of flower, with floral measurements.

Bivariate plot of the negative relationship between mean herkogamy and mean self seed by experimental population after 5 generations. Error bars are +/- 1 SEM. B1 and B2 are Bee populations, A1 and A2 are No Bee populations, and Source is the original population without treatment. Inset—cross-section of flower, with floral measurements.

Additionally, we identified a potential mechanism that may have contributed to the observed increases in selfing in the No Bee populations. Across populations, we found that selfing ability was correlated with herkogamy, the measured distance separating the male and female reproductive parts of a flower. Populations that had increased selfing rates had reduced herkogamy (see figure at left).

Our experimental evolution study demonstrates that rapid adaptation of reproductive strategy in response to pollinator loss is possible, but we also know that it comes with a cost. Selfing is complete inbreeding – and one disadvantage of inbreeding is a loss of genetic diversity, making populations vulnerable to extinction if further environmental perturbations occur.

Future directions

Current work on the monkeyflower mating system evolution project focuses on identifying specific regions of the genome that

evolve during a mating system transition from outcrossing to selfing. To accomplish this, we are developing a novel association technique, similar to standard marker-based QTL mapping, that targets highly divergent genomic regions from full-genome sequencing of experimental populations. Forthcoming results will help us understand how plant genomes respond to selection for increased selfing.

I originally became interested in mating system evolution during an NSF REU experience as an undergraduate, learning about northern cardinal (Cardinalis cardinalis) mate guarding and extra-pair paternity behavior. At MSU, I’m going back to my avian roots by developing a collaborative project with Dr. Tom Getty, graduate student Cara Krieg, and Dr. Lindsay Walters (Northern Kentucky University) to look at plastic responses of mating systems in the house wren (Troglodytes aedon).

For further information about these and other research projects, contact Sarah at bodbyl at msu dot edu or visit her website. Sarah also manages the NSF KBS GK-12 Bioenergy Sustainability project, a graduate training and research outreach program; check out their website here.

Posted in BEACON Researchers at Work | Tagged , , , , , | Leave a comment

This scientist's not-so-secret life

Posted on December 5, 2013 by Danielle Whittaker

Most of you know me by the name Danielle Whittaker. I am the managing director of BEACON, and also an animal behavior researcher. I’ve blogged about my research a couple of times: Deciphering avian aromas and The sweet smell of (reproductive) success.  

ChunkRockGirl1_Photo_by_Chris_SwitzerYou may not know that I have another name. I am the head referee for the Lansing Derby Vixens, and my derby name is Chunk Rock Girl. I have been skating for about 2 years, and refereeing women’s flat track roller derby for about a year and a half. Modern roller derby is a serious sport, and was even under consideration for the 2020 Olympics. I practice at least twice a week and referee derby bouts almost every weekend.

Recently, the PBS/NOVA web series “The Secret Life of Scientists and Engineers” invited me to appear on the show, to talk about science and about my “secret life” in roller derby. This Emmy-nominated series features all different kinds of scientists and the things we do when we’re not being scientists.

ChunkRockGirl2_Photo_by_Deanna_BradfordYou can check out my profile and watch my videos here, and learn about how I got involved in roller derby, what kind of work I do in the field, and the similarities between science and roller derby! 

The PBS/NOVA web series, “The Secret Life of Scientists,” provides a playful, humanizing snapshot of innovators who are shaping our world. Each episode features one of today’s leading scientists, and shows what happens when the lab coats come off—like a biologist who is also a professional wrestler, or a NASA astronomer who puts on fifty pounds of historical costuming to recreate Renaissance dances.  

Funded by the Alfred P. Sloan Foundation, this season’s “The Secret Life of Scientists and Engineers” will feature 16 scientists and engineers by the season’s end. 

Follow the Secret Life blog, and join the discussion on Facebook or Twitter @secretlifer. 

Posted in BEACON in the News, BEACON Researchers at Work | Tagged , , , , | Leave a comment

BEACON Researchers at Work: Evolution and Software Engineering

Posted on December 2, 2013 by Danielle Whittaker

This week’s BEACON Researchers at Work blog post is by MSU graduate student Erik Fredericks.

Erik FredericksI spent several years in the automotive industry as both a software developer and project manager, developing advanced systems that assisted in preventing accidents. These systems usually involved a camera and radar working together to sense obstacles or problems that a driver may face. That being said, I was indoctrinated in low-level, embedded C code that was very procedural and very straightforward. As more responsibilities piled on in my day to day activities, I was required to perform software testing, endure software quality reviews, and report on how “good” my team’s code actually was. What this all boils down to is the overall mysterious process known as software engineering.

At its core, software engineering is an approach for designing, developing, testing, and maintaining software. What makes it interesting from a research perspective is that it can be a quantifiable process. In essence, we can apply various techniques to optimize the different aspects of the software development cycle. This is where evolution, specifically evolutionary computation, can come into play, and is the focus of my research here at MSU with Dr. Betty Cheng.

Search-based software engineering (SBSE) is a growing field that applies search-based techniques to the engineering process. My research focuses on how evolutionary computation can be applied to topics within this field. As with other domains, evolution is typically used for optimization in SBSE. For instance, how can we enable software developers to reduce the amount of time spent creating test cases during development, so that they have time to spend on tasks that cannot be automated? Furthermore, how can we ensure that those test cases fully represent the spectrum of possibilities that the software may face while used by the general public?

Recently, I have been studying how complicated software systems perform over time, and particularly what conditions could cause them to react unexpectedly. Consider, for example, an autonomic robot vacuuming system, such as an iRobot Roomba. Its goal is to vacuum a specified area, accounting for problems such as obstacles, stairs, and the occasional pet. If the Roomba bumps into a wall or a table, it attempts to then find a new path that will be clear of problems and enable it to resume cleaning. If it encounters stairs, then it will turn around to ensure that it does not fall and damage itself. Pets, however, are allowed to ride for free.

The reason I bring up the Roomba is that it can be modeled as a dynamically adaptive system (DAS). A DAS is a fairly complicated software model consisting of different configurations. Each configuration can be selected at run time to mitigate a set of problems that are encountered. With a robotic vacuum, this can entail changing its mode of operation from cleaning in a straight line to slowly encircling a particularly dirty area. Each of these configuration modes is typically accompanied by a large set of parameters, each with different values to accommodate specific environmental conditions. As you can guess, this represents an explosion of possible states in which the robotic vacuum can operate, providing an ideal framework to explore with evolutionary computation!

Smart Vacuum System

Smart Vacuum System

To understand the different ways in which a robotic vacuum executes over time, I created a 3D simulation using the Open Dynamics Engine and a very nifty visualization engine from fellow BEACONite Jared Moore. Novelty search, an evolutionary technique that is concerned with finding a diverse representation of the solution space (rather than an optimal solution), was used to generate a set of system and environmental conditions that cause the vacuum to behave in many different ways. Furthermore, logging statements were inserted throughout the code that controlled the robot. The resulting set of logging statements, known as an execution trace, provides chronological information as to the steps taken throughout the simulation.

So what exactly does this tell us about how a vacuum performs over time, and particularly, why does this interest us? As a software engineer, any behavior that deviates from the expectation is typically bad. This deviation may demonstrate that there is a flaw in our software requirements, or perhaps in the system design, or even in the code itself. By having a representative, but more importantly diverse, set of execution traces, we can systematically localize software behavioral problems and determine an appropriate method for fixing the issue. More importantly, by using evolutionary computation, we have saved ourselves valuable time and effort in testing, validating, and finding new solutions not possible with traditional techniques!

[PDF] E. Fredericks, A. Ramirez, and B. Cheng, “Validating code-level behavior of dynamic adaptive systems in the face of uncertainty,” in Search based software engineering, Springer Berlin Heidelberg, 2013, vol. 8084, pp. 81-95.

For more information about Erik’s work, you can contact him at freder99 at msu dot edu.

Posted in BEACON Researchers at Work | Tagged , , , | Leave a comment

BEACON Researchers at Work: Students Become the Teachers – Teaching Evolution in the Classroom

Posted on November 25, 2013 by Danielle Whittaker

This week’s BEACON Researchers at Work post is by University of Texas postdoc Gwen Stovall. 

The Student Connection

Michael Ledbetter

Michael Ledbetter, lead evolution kit researcher

So, maybe the students didn’t have a good idea of “evolution” in the beginning. As far as they knew, they were unable to observe evolution… that takes eons to occur.  New experiences, though, offered new insights and the students didn’t retain these misperceptions for long.

In the early stages of developing an “evolution kit” for use as a high school demonstration, undergraduate students from the Freshman Research Initiative (FRI) were recruited to design, build, and implement the system. The FRI is a novel program comprised of more than 40% underrepresented groups and provides authentic research experiences to students. It also provided a partnership for the educational outreach missions of this project.

Both undergraduate and high-school students have been instrumental in all aspects of this project. From its first inception to the later stages, many high school and undergraduate students have contributed to these efforts, offering their unique perspective. Ultimately, a single undergraduate, Michael Ledbetter, stepped up to lead the team. Much of the success of this project is attributed to his steadfast, “heads down” approach, which culminated with conference speaking engagements and a recent publication Biochemistry and Molecular Biology Education (Ledbetter et al., 2013).

The Science Driver (Overview)

Underlying the kit’s development, we have merged previously developed technologies by combining continuous selection methods with a colorimetric reporter. We recreated one of Dr. Joyce’s continuous evolution systems using a modified variant of the Bartel self-ligating ribozyme (Wright and Joyce, 1997; Bartel and Szostak, 1993). The selection begins with a pool of T500 ligase ribozymes randomized at all 3 nucleobases composing the catalytic core (“T500 N3” pool). Serial dilutions cause selective pressure driving the evolution of species with improved catalytic function. With only 3 mutations at essential positions, improved catalytic function is evident after only one round of selection, thus suitable for the timeframe of a classroom period. This continuous evolution scheme is paired with a fluorescent reporter assay (strand displacement based). Using a fluorescent signal minimizes the equipment required to observe the selection. Therefore, the selection for ribozyme ligation and subsequent amplification is observed via simple colorimetric signal. This technological feat lends itself well to high school labs which lack more complex molecular biology equipment.

The Return (More specifically)

Eric Wei and Lea Drogalis (former high school students) performing preliminary kit tests

Eric Wei and Lea Drogalis (former high school students) performing preliminary kit tests

So, what are the students actually doing? They are observing the selection of a ribozyme. A ribozyme is a strand of RNA with catalytic function. In this case, the ribozyme can attach (or ligate) another RNA strand to its end (5’ end). This addition (containing a T7 promoter) permits the enzymatic amplification of the ligating strand. Thus, only the RNA variants with ligase activity are amplified. Simple dilution of the pool (with the reaction reagents) and a single round of amplification increases the concentration of “winning” ribozyme(s). Just 1-3 rounds of selection results with a fairly decent ribozyme.

What do the students actually see? Upon performing 1-3 rounds of ribozyme selection, the students observe the success of the reaction using a fluorescent reporter assay. Once the RNA strand (containing the T7 promoter) is ligated to the ribozyme, the end of the ribozyme displaces the paired fluorophore-quencher oligonucleotides, thus generating a fluorescent signal upon excitement with a UV light. Specifically, using the assay, the students test each round of their selection for ribozyme (or ligase) activity, which results with an increase in fluorescence over the course of the rounds.

What do the students actually get out of the experience? In addition to exposure to many of the techniques and theories of molecular biology, the students receive a working knowledge of catalytic RNA, which provides some insight into the RNA world hypothesis. Additionally, as is the focus of the demonstration, the students receive a first-hand demonstration of a selection for function, a necessary component of evolution.

The Mission

Shruti Singh presenting this project at a Rice University undergraduate conference.

Shruti Singh presenting this project at a Rice University undergraduate conference.

With the development of the first prototype of the “evolution kit” demonstration, we’re now focusing on testing, evaluation preparation, fine tuning the protocol, and getting it into the hands of high school students.

Shruti Singh, a former FRI student and current undergraduate mentor in the class, has now taken over the reins from Michael Ledbetter. Shruti, who recently presented the project at an undergraduate research conference at Rice University, is seeking to begin field-testing the prototypes with high school students soon. This will assess the feasibility of the required laboratory techniques for most high school students and if the tenets of this demonstration align with the educational curriculum. Most importantly, this will assess this experiment’s ability to introduce and reinforce fundamental evolutionary theory to high school students.

For more information about Gwen’s work, you can contact her at gwenstovall at gmail dot com.

Posted in BEACON Researchers at Work | Tagged , , , , , | Leave a comment

"The Man Who Bottled Evolution"

Posted on November 20, 2013 by Danielle Whittaker

A new Science paper from the Lenski lab is generating lots of buzz! Facebook’s popular page “I F*ing Love Science” sums up the Long-Term Evolution Experiment in this infographic:

Infographic

From the MSU press release:

There’s no peak in sight ­– fitness peak, that is – for the bacteria in Richard Lenski’s Michigan State University lab.

Lenski, MSU Hannah Distinguished Professor of Microbiology and Molecular Genetics, has been running his evolutionary bacteria experiment for 25 years, generating more than 58,000 generations. In a paper published in the current issue of Science, Michael Wiser, lead author and MSU zoology graduate student in Lenski’s lab, compares it to hiking.

“When hiking, it’s easy to start climbing toward what seems to be a peak, only to discover that the real peak is far off in the distance,” Wiser said. “Now imagine you’ve been climbing for 25 years, and you’re still nowhere near the peak.”

Only the peaks aren’t mountains. They are what biologists call fitness peaks – when a population finds just the right set of mutations, so it can’t get any better. Any new mutation that comes along will send things downhill.

The bacteria in Lenski’s lab are still becoming more fit even after a quarter century, living in the same, simple environment.

Biologists have known that organisms keep evolving if the environment keeps changing, but they’ve previously thought that adaptation would eventually grind to a halt if the environment stayed constant for a long time.

Wiser pulled hundreds of samples from the deep freezer that contains a frozen fossil record – bacteria all the way back to generation 0 in Lenski’s 25-year experiment. And these fossils, unlike dinosaurs, are alive. So they can be competed against samples from different generations to measure the trajectory – the path – of the bacteria as they climbed for 58,000 generations toward the fitness peaks.

“There doesn’t seem to be any end in sight,” Lenski said. “We used to think the bacteria’s fitness was leveling off, but now we see it’s slowing down but not really leveling off.”

Wiser found that the trajectories matched a type of mathematical function called a power law. Although the slope of the power-law function gets less and less steep over time, it never reaches a peak.

Noah Ribeck, co-author and MSU postdoctoral researcher, built a model using a few well-understood principles.

“It was surprising to me that a simple theory can describe the entirety of a long evolutionary trajectory that includes initially fast and furious adaptation that later slowed to a crawl,” Ribeck said. “It’s encouraging that despite all the complications inherent to biological systems, they are governed by general principles that can be described quantitatively.”

When will it end?

“I call this the experiment that keeps on giving,” Lenski said. “Even after 25 years, it’s still generating new and exciting discoveries. From the models, we can predict how things will evolve – how fit the bacteria will become – if future generations of scientists continue the experiment long after I’m gone.”

Lenski hopes that an endowment could be secured to keep the experiment going forever, he added.

Lenski’s research is supported in part by the National Science Foundation. Wiser, Ribeck and Lenski are also participants in the NSF-funded MSU BEACON Center for the Study of Evolution in Action.

———

M. J. Wiser, N. Ribeck & R. E. Lenski, 2013. Long-term dynamics of adaptation in asexual populations, Science.

See also:

 

Posted in BEACON in the News | Tagged , , | Leave a comment