Evaluating the contribution of symbiotic microbes to reproductive isolation of vertebrate populations

Personnel: Future Postdoc, Kevin R Theis, Danielle J. Whittaker
Keywords: Behavior, Hologenomics, Speciation, Vertebrates, animal behavior, microbiome
Abstract: Microbes profoundly influence the biology and evolution of their animal hosts, potentially including the process of speciation itself. For example, bacteria can contribute to the production of chemical signals that function as pre-mating reproductive barriers and reinforce evolutionary divergence. We will examine this process in multiple subspecies of a rapidly evolving songbird, the dark-eyed junco (Junco hyemalis) using chemical and microbial ecology approaches. Our previous work has demonstrated the importance of chemical signals in reproductive behavior in this species, and that microbes associated with the preen gland likely contribute to these signals. In Aim #1, we will experimentally test whether chemical signals reflect social relationships via shared symbiotic microbes, and compare the effects of sex and subspecies on microbe flow. The goal of Aim #2 is to test whether chemical signals reflect major histocompatibility complex (MHC) genotype. This project represents the first attempt to examine the interaction among vertebrate social behavior, host genotype, and the symbiotic microbes that affect phenotypes important in reproduction and, ultimately, speciation. The proposed research will bring together studies of host social behavior, genotype, and environment with symbiotic microbial and chemical ecology in the context of rapidly diverging vertebrate populations. This project will make a major contribution to the developing paradigm shift that is functionally redefining individuality to include symbiotic microbial communities.

BEACON Native American/Alaskan Native Initiative (NAANI)

Personnel: Judi Brown Clarke, Wendy F. Smythe
Keywords: Cultural Competency, Data Nuggets, Diversity
Abstract: (No abstract entered)

Field Research Experiences for Early Career Undergraduates at KBS – Summer 2018

Personnel: Judi Brown Clarke, Gregory Goins, Katherine Gross, Future Undergraduate Student, Danielle Zoellner
Keywords: Diversity Recruiting, Eco-evolutionary processes, Ecological Interaction, Undergraduate Education
Abstract: This proposal is to fund four students to participate in the Undergraduate Research Apprentice (URA) program at KBS to increase their understanding ‘evolution in action’ and BEACON related research. The URA program complements the NSF funded REU program at KBS by focusing on early career and under-prepared students and introduces them to skills needed to be successful in further research experiences by increasing their understanding of evolution and the nature of scientific research. BEACON fundsleverages support from other MSU sources for the KBS URA program. URAs gain research experience by taking courses at KBS (tuition funded separately) and assisting a graduate student or postdoctoral mentor with their research. URAs also participate in a professional development program that focuses on skills needed to pursue STEM careers and graduate study. Students gain experience in science communication by presenting a poster or writing a blog (posted at BEACON and KBS) summarizing their research. Funding is also requested for URAs to present their research at regional and national meetings. The program addresses diversity and educational goals of BEACON by targeting recruitment of under-represented minorities and first-generation students and preparing them for STEM careers. We work with minority-serving programs at MSU and partner schools to identify and recruit highly motivated students to the URA and REU program. Evaluation of the program is done by post-program surveys of the participants (students and mentors) using surveys developed by the NSF.

Tradeoffs During Niche Adaptation Stabilize Niche Construction

Personnel: Brian Connelly, Ben Kerr, Joshua Richard Nahum, Undrgraduate Student, Katrina van Raay
Keywords: Agent Based Modeling , Cooperation, Eco-evolutionary processes, niche construction, tradeoff
Abstract: Through just about all that they do, organisms change their environment. From metabolizing resources to building dams, these niche constructing behaviors can profoundly influence natural selection in both positive and negative ways. This project will develop and test a new theory that as populations modify their niche, tradeoffs can emerge that enable positive niche constructing behaviors to be maintained by a paper-rock-scissors dynamic. We will also extend our model to include epistasis, which may lead populations to become “addicted” to niche construction as they continue to adapt to their environment.

Evolution of the gut microbiome: a long-term analysis in a wild mammal population

Personnel: Kay E. Holekamp, Connie A. Rojas, Unnamed Undergraduate Student, Kevin R Theis
Keywords: Mammal, animal behavior, bacteria, microbiome, social networks
Abstract: Despite much recent debate over the influence of the microbiome on host evolution, the potential impact of host-afforded niche spaces on the evolution of microbiomes remains largely unconsidered. We aim to identify the environmental and host characteristics driving the evolution of animal microbiomes. We will evaluate whether functional redundancy is maintained over evolutionary time in the gut microbiome of the spotted hyena (Crocuta crocuta), as indicated by consistent patterns in gene representation despite fluctuations in the microbiota, across 30 yrs of study constituting tens of thousands of microbial generations. Our retrospective longitudinal investigation will determine how much the hyena gut microbiota and microbiome change within individual hosts across decades, and within a single population across generations, by studying the gut microbes of 12 adult females from four matrilines, each spanning three generations. We will also determine the extent to which there have been alterations in the genomes of prominent Bacteroides strains within the hyena gut over time and across generations. We will determine which, if any, environmental and host characteristics are associated with longitudinal variation at microbiota, microbiome, and bacterial strain levels. By integrating data from host, microbiota, microbiome, and specific bacterial strain profiles, and capitalizing on intensive computational tools to elucidate interactions between hosts and the microbiome, this project will elucidate principles governing host microbiome assembly, functional capacity and evolution.

Evolutionary Optimization of Water and Nutrient Use Efficiency for Sustainable Agricultural Intensification

Personnel: Mohammad Abouali, Kalyanmoy Deb, Student Graduate, A. Pouyan Nejadhashemi, Proteek Chandan Roy
Keywords: Applied Evolution, Biologically Inspired Computing, EMOO, Evolutionary Algorithm, Evolutionary Multi-objective Optimization, Genetic Algorithm, tradeoff
Abstract: One of the major challenges that the world is facing in coming decades is how to meet the growing food demand without compromising the integrity of our environment. An estimate suggests that global food production needs to be increased by 60-110% between 2005 and 2050. By closing the yield gaps, which is the difference between attainable yield and actual yield in a region, most of the countries are expected to meet food self-sufficiency or improve their current food self-sufficiency levels. Water and nutrient availability are the major production limiting abiotic factors in the regions where the yield gaps are high. In this regard, effective water
and nutrient management plays a crucial role in food security by closing the yield gaps. Optimizing water and nutrient management not only improves crop yield, but also reduces production cost, conserves resources, and protects our environment. Due to the presence of multiple conflicting objectives, computationally expensive simulation routines, and nonlinearities in objectives and constraints, evolutionary computation methods with their recent advances remain as potential tools for addressing these important socio-economic problems. In this project, we combine the expertise of Nejadhashemi in modelling of crop growth through water and nutrient usage and Deb’s evolutionary many-objective optimization research together to address this important social issue. DSSAT software will be combined with recently proposed U-NSGA-III procedure for arriving at important solution principles.

The role of individual intelligence in the evolution of collective sensing and defense

Personnel: Andrew S. Gersick, Kay E. Holekamp, Frants Jensen, Ari Strandburg-Peshkin, Future Undergraduate Student
Keywords: Behavior, Communication, Cooperation, Intelligence, Social Evolution, collective behavior
Abstract: Cooperation across dispersed, heterogeneous social networks is a keystone aspect of human society, yet its origins are unclear. Recent research has shown that simple individual rules can drive coordinated movement in dense, homogenous groups such as fish schools. Far less understood are the mechanisms of coordination under conditions of social and cognitive complexity, as in stable societies with differentiated relationships mediated by long-term memory. Like humans, spotted hyenas hold a dominant ecological niche largely via collective action. Hyenas disperse widely to forage in small subgroups, but clans converge rapidly to defend prey and territory against lions and rival groups. Lone hyenas recruit allies using long-distance “whoop” calls, allowing clans to collectively detect threats. This system depends on enough individuals responding to these calls, but not all hyenas do. By assessing how hyenas “decide” whether to opt into collective action, we aim to illuminate the role of two aspects of intelligence (recognition of clan-mates and memory of past events) in shaping this behavior. Combining field experiments with novel tags that continuously record movements and vocalizations, we will investigate hyenas’ recruitment system on two timescales: immediate responses to signals, and long-term ranging patterns following exposure to threats. We will then use movement and evolutionary models to link individual decision-making to collective outcomes, and to investigate the evolutionary history of individual cognition supporting collective action in a complex social species.

Homage to Robert Johnson: Stab them, poison them, or both?

Personnel: Shyam Aravamudhan, NCA&T Graduate Student, Joseph L Graves, Michael Maiden, Chris Waters
Keywords: Adaptation, Antibiotic Resistance, Biofilms, bacteria
Abstract: Multidrug resistant bacteria are ubiquitous. Attempting to poison them has only increased their spread. This proposal addresses whether adding an additional method, physical disruption will deter the evolution of resistance. We will utilize thermoplastic fused deposition modeling (FDM) 3D printing to fabricate sharp edged nanostructures composed of toxic metals (e.g. silver and iron oxide nanoparticles) to test the evolvability of resistance to poisoning, stabbing, or both on these surfaces. We will evaluate the physical characteristics of resistant clones and their genomic architecture should such emerge, as well as test the sustainability of these nano surfaces as antimicrobials.

BEACON TEACHERs Researching EVOLUTION (TEACHER REVOLUTION)_National

Personnel: Katie Dickinson, Kara Haas, Louise Souther Mead, Christine Parent, Tessa Solomon-Lane, Future Undergraduate Student, Jack Sullivan, Luis Zaman
Keywords: Informal Education, K-12 Education, Professional Development, Public Outreach
Abstract: The National BEACON TEACHER REVOLUTION Program (TEACHERs Researching EVOLUTION) will provide current K-12 teachers with a summer research experience in a mentored lab while simultaneously offering a professional development experience that identifies effective ways to integrate their research experience and science practices into their classroom. Teachers will have the opportunity to carry out original research, under the guidance scientists mentors at our consortium institutions. Specifically, we propose to place two teachers each at University of Washington, University of Idaho, University of Texas, and Michigan State University. A networked learning community will be established via weekly professional development via videoconferencing. This program builds on successful, evidence-based, programs that show research experiences positively impact the teachers as well as their students. Teachers will expand their understanding of evolutionary biology and the nature of science. They will impact their research lab by offering new perspectives and creating lesson plans and web-based materials to share their experience with a broad audience. BEACON’s legacy of exemplary outreach with K-12 teachers will continue and expand with this successful, national Research Experience for Teachers program

BEACON and Beyond: Broadening Participation and the Scope of Public

Personnel: Travis Hagey, Hans A. Hofmann, Tessa Solomon-Lane, Alexa Warwick
Keywords: Graduate Education, K-12 Education, Postdoc Education, Public Outreach, Science Education
Abstract: Engaging with the public is a fundamental professional responsibility for academic scientists, and it is required by many funding agencies. Engagement is also a civic responsibility. Although Americans have confidence in science and trust in scientists, science denial and the spread of false claims, especially in politics, can be dangerous for society. However, participation in public engagement is often limited to “the dedicated few,” as many are unaware that public engagement benefits professional development and scholarship. We have developed an innovative, evidence-based program that motivates, recruits, and trains graduate students and postdocs in best practices for public engagement and pairs them with opportunities to engage. We aim to professionally package and disseminate this program so it can be replicated in any institution to concurrently enhance trainee professional development and public engagement. We will also expand our program to include a module on framing a scientific audience for different audiences, one of the most important communication skills. Finally, although BEACONites currently engage the public in a wide variety of ways, our survey data show striking underrepresentation of engagement with public policy and policy makers. We will develop a specialized curriculum for engaging this audience and facilitate policy engagement opportunities. This work adds new and important dimensions to public engagement training and action and will reshape how scientists learn to communicate to diverse and influential audiences.

Decoy mutations: engineering fitness landscapes to delay unwanted evolution

Personnel: Christoph Adami, Jeffrey E Barrick, Future Graduate Student, Thomas LaBar, Luis Zaman
Keywords:
Abstract: Synthetic biology is rapidly advancing our ability to create complex genetic circuits and perform feats of metabolic engineering. This progress is enabling medical, manufacturing, and energy breakthroughs. However, when organisms are engineered to perform some desired task, it is typically at the expense of resources that would otherwise be invested in producing more offspring. For this
reason, evolution often favors organisms that have accumulated mutations disrupting the engineered function. Here, we propose to harness clonal interference, a phenomenon that occurs in large asexual populations whereby competition between multiple beneficial mutations slows the fixation of any one allele, as a means of prolonging the lifetime of a burdensome function in a population. The basic setup is to add the possibility for one or more decoy mutations into the fitness landscape of a cell. The decoy mutations are designed to “distract” adaptive evolution from inactivating the engineered function. Our two key questions for this proof-of-principle seed project are: (1) What is the optimal design for a suite of decoy mutations to maximally extend the lifetime of the desired function in a population? (2) Can we extend the half-life of an engineered trait even further by dynamically altering the environment in a way that replenishes the supply of decoy mutations? We will examine both questions using a BEACON-inspired combination of modeling approaches and microbial evolution experiments.

A Novel Phylogenomic Framework for Forensic DNA Localization

Personnel: Gregory Bonito, Student Graduate, Kevin Liu, Wei Wang
Keywords: Localization, Phylogenomics, Phylogeography
Abstract: Forensic DNA analysis is indispensable to criminal investigation and the criminal justice system. The goal of our interdisciplinary project is to create and validate a novel tool for localizing DNA samples, including trace evidence from non-human organisms. Our computational tool will provide important investigatory clues, especially in international cases (e.g., “hit-and-run” terrorism using air travel), and criminal evidence.

The key insight is that evolutionary relatedness between a forensic DNA sample and population-based DNA samples from a particular locale is a telltale signature of local origin. Our phylogenomic pipeline therefore incorporates FastNet, our recently developed computational method for fast and accurate phylogenomic inference, for accurate reconstruction of evolutionary relationships between forensic and population-based DNA samples. Statistical approaches will be used to test for significant sample enrichment from a single location of origin. This yields greater forensic power since the probability that multiple independent samples originate from the same location by chance becomes vanishingly small as more samples are considered. We will perform an empirical cross-validation study to demonstrate the power of our tool in comparison to the current state-of-the-art. The study provides a proof-of-concept using genomic DNA sampled from globally distributed fungal populations. More broadly, we intend to demonstrate the power that phylogenomics and evolutionary thinking can bring to forensic DNA analysis.

EXPLORING NEW APPROACHES TO ENSEMBLE AND NEURAL NETWORK TOPOLOGICAL EVOLUTION FOR TOUCH-BASED AUTHENTICATION

Personnel: NCA&T Graduate Student, Sandeep Kulkarni, Kaushik Roy
Keywords: Biometrics, Machine Learning
Abstract: In a typical access-control scenario, once authentication has taken place, a session remains active and a device remains unlocked until an individual or the server closes the session. In the event that a user leaves their device or workstation without closing an active session, the system may be accessed by an unauthorized party during the period before session timeout. To counteract this, active authentication can be implemented to continuously monitor the identity of the individual using a device. Behavioral biometrics can be monitored unobtrusively in the background, without the need to explicitly query the user for input. In this work, we will use touch-screen swipe pattern authentication to avoid smart-phone session hijacking.
This work will proceed in three phases. First is to collect data and establish a set of baseline classifiers for performance comparison. These will include a traditional feature-distance-based classifier, a Convolutional Neural Network (CNN), and a Recurrent Neural Network (RNN). Phase two will convert the feature-based classifiers from phase 1 into neural networks, and use them as a seed population to evolve an optimum ensemble using diversity as part of the fitness function. We will then apply stacked-generalization in an effort to further improve ensemble accuracy. Phase three will explore the use of a learned fitness function to accelerate the fitness evaluation in the evolution of domain-specific deep-nets.

It only gets worse from here: adaptation in the face of ever-larger pulses of stress

Personnel: Jannell Bazurto, Emily Louise Dolson, Ben Kerr, Christopher J Marx, Sonia Singhal
Keywords: Adaptation
Abstract: Lowering rates of environmental change can expand possible evolutionary trajectories (Lindsey et al. 2013). We propose to evolve Methylobacterium extorquens for formaldehyde growth using varying rates of change in formaldehyde concentration, and to use an NK model to characterize evolution on fitness landscapes that differ between growth and survivability. In the Marx lab, we previously evolved M. extorquens PA1 experimentally to grow on formaldehyde using step-wise increments of near-lethal doses of formaldehyde. The resulting populations showed that growth on formaldehyde could be achieve by single mutations that activated uncharacterized stress response systems and/or rescued M. extorquens from extinction. Here, we propose to use selective regimes that go through sub inhibitory and sub lethal incremental steps to enrich for more paths toward formaldehyde growth. Lower rates of change may also provide an opportunity for improved formaldehyde utilization (intrinsically linked to detoxification) to be a significant selective pressure, which may be in direct competition with survivability. To inform and extend the results from M. extorquens, the Kerr and Ofria labs will implement an NK model that incorporates varying rates of environmental change and imposes different selective pressures during survival and growth. This combination of empirical and computational work will inform our understanding of how and why organisms adapt differently depending on the rate of environmental change.

Deciphering Innovative Free-form Solution Principles Using Genetic Programming and Multi-objective Optimization

Personnel: Wolfgang Banzhaf, Kalyanmoy Deb, Abhinav Gaur
Keywords: Genetic Programming, Innovization, Multiobjective Optimization
Abstract: In scientific and technical problem solving activities, we are often interested to learning underlying principles for constructing a high-performing solution. In game-playing activities, this is precisely the approach, we often follow. By playing the game many times against various scenarios, we gather “rules” that are found to produce successful outcomes. In future events, we try to use these rules to make our moves. In recent studies in engineering design and other optimization problems, Deb suggested a generic two-step “innovization” method to arrive at innovative solution principles through optimization in the hope of deciphering useful design principles for a better understanding of the problem: a) A multi-objective to produce multiple trade-off solutions, b) A machine learning based feature extraction procedure to decipher hidden design principles involving variables, and objectives. Although most innovization studies have been confined to manual pair-wise regression analysis, automating the innovization process is the next challenge. Also, manual studies considered design principles (or rules) to have a specific mathematical structure, but such a restriction not only limits its use in generic problems, it does not enable discovery of other important types of relationships pertinent to a problem. We plan to use genetic programming (GP) framework for discovering free-form rules hidden in a set of trade-off optimized solutions. Then, we plan to investigate the use of genetic programming (GP) in more advanced innovization tasks – dimensionally meaningful rule using strongly-typed GP, lower-level and temporal innovization for discovering different rules. Methods will be tried on engineering problems and will be compared with manual regression and other standard method

Experimental evolution of microbiomes for ‘studs’ and ‘duds’ in Drosophila melanogaster

Personnel: Future Graduate Student, Chad Christopher Smith, Danielle J. Whittaker
Keywords: Communication, Evolution In Action, Experimental Evolution, Mate Choice, Metagenomics, Mutualism, Sexual Selection, animal behavior, bacteria
Abstract: Animals are host to diverse communities of microbial symbionts that have played an integral role in their ecology, evolution, and behavior. Archie & Theis identified sexual selection as one of three most promising areas at the interface between animal behavior and microbial ecology, for at least two reasons. First, sexually selected traits such as ornaments and weapons are highly condition-dependent, and the microbiome is central to nutrition and immune function during development. Second, microbes influence animal communication by contributing to the production of signals by metabolizing host substrates, particularly in olfactory communication. We propose to test whether ecological and evolutionary changes in microbiomes affect male sexual attractiveness in Drosophila melanogaster using ‘one-sided’ host-mediated selection on microbiomes. We will produce ‘stud’, ‘dud’, and ‘random control’ microbiome selection lines by propagating the microbiomes of attractive, unattractive males, and random males over five selection cycles against a static host genetic background. Illumina sequencing of 16S rRNA amplicons will be used to determine whether selection lines differ in microbiome structure and to understand how changes in community states vary over the course of the experiment. To assess the proximate cause of variation in attractiveness between treatments, courtship rate and cuticular hydrocarbons will be measured to determine whether microbiomes affect sexual behavior and olfactory signals important in sexual selection, which has previously been suggested in D. melanogaster. We will also measure several life history variables to determine whether microbiomes influence covariation with sexually selected traits.

Off the deep end of the gene pool: Deep-sequencing and metabolomics analysis as a framework for uncovering fitness
landscapes and modeling evolvability in engineered metabolic networks

Personnel: Eric Bruger, Student Graduate, Cecilia Martinez, Christopher J Marx, Sergey Stolyar
Keywords: Adaptation, Bacterial Populations, Biotechnology, Epistasis, Evolvability, Fitness Landscapes, Gene Networks, Mathematical Modeling, Synthetic Biology, systems biology
Abstract: Evolution in many systems is fraught with constraints that limit the paths that are adaptive. Many combinations of genetic changes result in nonlinear fitness effects, or epistasis. Epistasis can potentially reduce (magnitude epistasis) or in extreme cases reverse (sign epistasis) the combined fitness between alleles. From the perspective of an adaptive landscape, epistasis adds ruggedness, making it more difficult for populations to traverse the landscape to effectively search and find fitness peaks. This problem presents itself in an system with sufficient epistasis, including both naturally evolving populations as well as attempts to optimally engineer organisms for industrial applications. Epistasis can be identified at many levels of organization, and is very likely to occur within metabolic pathways. We propose to study epistasis at an unprecedented level of resolution for a metabolic network by adopting a new method to generate large degrees of genotypic variation on short time scales and assessing changes in pooled populations through deep-sequencing. Combined short- and long-read sequencing results will provide large amounts of genotype-fitness data, which we will choose among to pursue more in-depth phenotypic characterization by measuring enzyme and metabolite levels, and fitting mathematical models of the entire network. The result of this combined approach will be a better understanding of epistasis in this metabolic system and an overall approach to identifying and addressing epistatic interactions broadly.

THIRD YEAR: BEACON REU Program @ MSU

Personnel: Judi Brown Clarke, Multiple Undergraduate Students
Keywords: Diversity, Diversity Recruiting, Responsible Conduct Of Research
Abstract: BEACON’s Summer REU Program @ MSU is a 10-week intensive residential program targeting the recruitment of a diverse group of students to conduct research with faculty, graduate students and post-doc mentors. The settings included wet & dry labs, computer/simulations, field experiences, and/or a combination. Over the past five summers, MSU BEACON has served 269 undergraduate students (freshman to 5th-year seniors) and spent over $1 million by leveraging and blending funding from several sources. A distinguishing feature of BEACON’s summer research program is that before the REU program begins, and in regular meetings during the summer, research mentors received formal training in mentoring from BEACON’s Diversity Director, Judi Brown Clarke, who is available to both mentors and mentees over the course of the summer to help solve problems that arise, as well as to enhance the mentors’ professional development and cultural competency. The long-term goal of the program is to train graduate students and postdocs to build mentoring relationships that can be used to increase recruitment and retention, share cultural and organizational knowledge, and help individuals achieve personal and professional objectives. The short-term goals are to ensure that undergraduate interns have an explicit mentor to help guide their research and laboratory training, and that the graduate and postdoctoral mentors have the formal skills and resources necessary to serve as effective mentors.

A one-two punch: combining killing systems within a single cell by going “pro”, and the mystery of the one-punch

Personnel: Joshua Richard Nahum, Future Undergraduate Student, Katrina van Raay
Keywords: Agent Based Modeling , Eco-evolutionary processes, Experimental Evolution
Abstract: Chemical warfare in the microbial world is ubiquitous. One kind of chemical weapon microbes employ is a proteinaceous toxin called a bacteriocin. The best-studied bacteriocins are the colicins, produced by and active against Escherichia coli. Many colicin systems encode suicidal lysis genes, such that the producing cell releases the toxin through cell lysis. Released colicin kills sensitive competing cells allowing immune clones of the producer to capitalize on the liberated resources. However a major group of colicin systems lack this lysis gene, and it has been unclear how such toxins were released from the producing cell. Here, we explore a newly discovered union between such a colicin system and a prophage (bacterial virus that has been incorporated into the bacterium’s genome), where release of the colicin occurs via phage-encoded cell lysis. This union should be detrimental to susceptible cells, which would be hit by a one-two punch: once by the colicin, once by the phage. Preliminary mathematical modeling suggests that the success of this dual killing system depends on the prophage’s ability to produce infectious virions. If the prophage is cryptic (encoding lysis, but not producing infectious phage), then this dual system is susceptible to invasion by “cheaters” that only possess the colicin system (and its immunity) but do not lyse. Here, we explore this social dilemma and its potential resolution in detail through a combination of mathematical modeling, microbial evolution experiments, and agent-based simulations.

Harnessing Eco-Evolutionary Dynamics for Open-Ended Evolution of Intelligence

Personnel: Wolfgang Banzhaf, Emily Louise Dolson, Sarah Elizabeth Johanknecht, Steven Jorgensen, Charles Ofria, Future Undergraduate Student, Michael Wiser
Keywords: Digital Games, Eco-evolutionary processes, Evolutionary Algorithm
Abstract: When trying to solve complex problems, we often have intuition about the building blocks needed to find a solution. However, it is challenging to transfer this intuition to an evolutionary algorithm without restricting or misleading evolution. Our previous research on Eco-EA, an evolutionary algorithm that uses ecological dynamics to maintain diverse populations, provides a potential solution. By associating limited resources with tasks that may be useful as “hints” to solving a larger problem, we can create a diverse ecology of partial solutions, wherein only a subset of the community will follow each hint. If a hint proves to be incorrect, few lineages will follow it to an evolutionary dead end. We propose to expand this approach and test its effectiveness in a complex domain: evolving an intelligent board-game playing agent for The Settlers of Catan. In addition to being a complex and multi-faceted problem, Catan is a popular game, making it an excellent vehicle for outreach. As people play, we will associate limited resources with the ability to predict each player’s next move. Thus, player strategies will automatically translate to hints for the algorithm, producing many co-existing niches and resulting in a crowd-sourced AI. This project will unite ideas from evolution, ecology, and artificial life to advance the state of the art of evolutionary computation and evolved artificial intelligence. At the same time, it will provide exciting outreach activities, giving the public hands-on experience with an application of evolutionary and ecological theory.

Scientific Virtues Toolbox: Meeting RCR Needs for BEACON – And Beyond!

Personnel: Eric Berling, Robert Heckendorn, Michael O’Rourke, Robert T. Pennock
Keywords: Responsible Conduct Of Research
Abstract: The Scientific Virtues Toolbox will develop responsible conduct of research modules and conduct workshops that will be available for all BEACON participants for Friday seminar and Congress. It will collect pilot data to be used for external grant opportunities that we have already identified.

SECOND YEAR: Integrating Research, Diversity, and Education into the Long-Term Evolution Experiment

Personnel: Jason Nyerere Bundy, Kyle Joseph Card, Nkrumah Grant, Richard Lenski
Keywords: Adaptation, Diversity, Experimental Evolution
Abstract: The Long-Term Evolution Experiment (LTEE) with E. coli has been running for almost 28 years and over 64,000 generations. It is an exemplar of evolution-in-action studies and one of the cornerstones of BEACON. In keeping with the goals of BEACON, I have sought to ensure that the LTEE contributes not only to the research mission of BEACON, but also to achieving BEACON’s aspirations for diversity and education. To those ends, I have recently recruited three diverse doctoral students (including two from underrepresented minorities, and two with disabilities), all of whom are engaged in research on the LTEE and active in BEACON. I have also hosted – and would like to host again – two high school biology teachers to design microbial evolution-in-action experiments for their students. To assist me with those aims, I request half-time support for the three doctoral students, and summer support for the two high-school teachers, over each of the next two years. The doctoral students are examining (i) historical contingency including its importance for adaptive radiations in new resource environment; (ii) the effect of epistasis on the emergence of antibiotic resistance and its fitness costs; and (iii) phenotypic plasticity and changes in fitness of the LTEE lines when presented with a changed physical environment. The high-school teachers will develop and refine procedures for evolution-in-action experiments with bacteria that they can use in their classrooms and also disseminate more broadly.

SECOND YEAR: Modeling the Evolution and De-Evolution of Biological Complexity

Personnel: Christoph Adami, Thomas LaBar, Richard Lenski, Dariya K. Sydykova, Claus O. Wilke
Keywords: Adaptive Complexity, Digital Evolution, Experimental Evolution, Mathematical Modeling
Abstract: The mathematical description of evolving living systems has made great strides in the last century, but theory is concerned mostly with understanding evolution on short time scales: those relevant for the fixation of a single or perhaps a pair of loci. Modern developments in experimental evolution have highlighted a missing topic in the mathematics of population genetics, namely a theory that describes how populations adapt over the long term. A recent theory by Wiser Ribeck, and Lenski (WRL theory) has shown that evolution over the long term can be predicted given a sufficiently long trajectory in the past. It turns out that any theory that can predict forward evolution, should also be able to predict the loss of fitness over time, due to drift (for example when population sizes are small. We can use such a theoretical prediction to estimate the complexity of the organism that is losing fitness, as several theoretical approaches suggest that the rate of fitness loss is determined by the complexity of the initial organism. We will develop the theory to estimate the fitness loss over long-term, and validate it by performing drift-response experiments with well-characterized genomes of the Long-Term Evolution Experiment (LTEE). If successful, we will have a new quantitative way to link fitness to organismal complexity.

SECOND YEAR: Gene-level selection, phage, and the evolution of antibiotic resistance

Personnel: Alita Burmeister, Jorden Schossau, Jim Smith, Paul E Turner
Keywords: Antibiotic Resistance, Computational Evolution, Experimental Evolution, H
orizontal Gene Transfer
Abstract: Gene-level selection appears to be common in nature, where it underlies the spread of factors important to human health, such as antibiotic resistance genes and pathogenicity islands. However, the evolutionary and ecological mechanisms underlying gene-specific replication in natural populations remain relatively understudied and are not well understood. Classic studies in microbial genetics have determined that replication of individual genes can occur through horizontal gene transfer from one genome to another. One of the major modes of horizontal gene transfer involves the packaging of host DNA into phage capsids, which deliver the DNA to another cell where it recombines into the chromosome. This process, termed transduction, has long been used to transfer genes in laboratory molecular biology. Our BEACON research will identify basic principles of how Salmonella phage transduction evolves in natural populations, laboratory populations, and in theoretical computational populations. To do this, we will use a complementary set of approaches including molecular phylogenetics, experimental evolution, molecular biology, and evolutionary computation. Knowledge of these basic principles will allow the prediction of future pathogen evolution, and inform policy and best practices for long-term effectiveness of antimicrobial drugs and therapeutic phages.

Computational studies to elucidate evolutionary conservation of phosphorylation sites

Personnel: NCA&T Graduate Student, Student Graduate, Dukka KC, Robert Newman, Claus O. Wilke
Keywords: Computational Biology
Abstract: Phosphorylation is a key posttranslational modification that controls the regulation of nearly all cellular processes. Importantly, because cellular phosphorylation networks play a central role in regulating gene expression patterns and cellular metabolism, re-organization of these networks is likely to lead to systems-level changes in cellular physiology that can impact evolutionary parameters in a manner analogous (or complementary) to genetic variation. Interestingly, though phosphorylation-dependent signaling is a highly conserved mode of information processing across eukaryotic species, recent studies suggest that individual phosphorylation sites on orthologous proteins can vary substantially across species. While many studies have focused on individual phosphorylation sites, there are relatively few that have explored evolutionary conservation of phosphosites across various species.
To perform a systematic comparison of phosphorylation network between species, we first must expand phosphosite datasets for various organisms. We must then develop an approach to compare phosphorylation sites across species. Thus, in this pilot project, we will pursue two primary objectives: First, we will build on our previous studies to develop a generalizable approach for the development of phosphorylation site prediction tools that utilize machine learning approaches. This will ultimately allow us to enrich the information about phosphosites in various organisms. Second, we will develop an approach to perform comparative analyses of phosphorylation sites across species. These algorithms & datasets will be useful for exploring the evolutionary conservation of phosphorylation sites across species and for understanding the rules underlying phosphorylation-dependent signaling.

Microbes and Microplastics: Is there evidence for selection in a novel deep ocean niche?

Personnel: BEACON funded undergraduate researcher, James Arthur Foster, NCA&T Graduate Student, Joseph L Graves, Scott H. Harrison, Eric Mittelstaedt, Janet Williams
Keywords: Bacterial Populations, Biological Evolution
Abstract: Plastics are a serious threat to ocean ecosystems. Plastics are bio- and photo-degraded into tiny fragments that are being found in amounts on the deep-sea sediments of the ocean floor comparable to levels found in sediments of shallow and intertidal sea zones. Recent studies have identified plastics to be almost ubiquitous across samples assayed in several deep-sea environments including the North Atlantic, the Southern Ocean, the Gulf of Guinea, and the Mediterranean Sea (Van Cauwenberghe et al., 2013; Woodall et al., 2014). Another study has examined how this “plastic rain” effects the ocean floor ecosystem, finding that six of nine animals that were examined from diverse parts of earth’s oceans, have microfiber plastics in their systems (Taylor et al., 2016). It is reasonable therefore to suppose that there are recent (in evolutionary terms) emergence of niches for plastic-degrading bacteria (e.g., Zettler et al., 2013) in deep ocean sediment that would lead to selection for those bacteria. One of us (Mittelstaedt) recently collected ocean sediment cores from the Mid-Atlantic ridge (13°N -14°N). We propose to quantify changes in microbial communities’ composition and strength of selection across sediment locations and layers, and to examine plastic fragments directly for bacterial inhabitants. In short, we study evolution in action, by comparing pre-plastic and no plastic sediment layers and locations respectively, to how other parts of the deep ocean sediment microbiome have been impacted by the presence of plastic. This occurs in conjunction with our study of a unique and largely unexplored plastic cycle from the Mid-Atlantic surface to the sea floor.

Homage to Robert Johnson: Stab them, poison them, or both?

Personnel: Shyam Aravamudhan, NCA&T Graduate Student, Joseph L Graves, Michael Maiden, Chris Waters
Keywords: Adaptation, Antibiotic Resistance, Biofilms, bacteria
Abstract: Multidrug resistant bacteria are ubiquitous. Attempting to poison them has only increased their spread. This proposal addresses whether adding an additional method, physical disruption will deter the evolution of resistance. We will utilize thermoplastic fused deposition modeling (FDM) 3D printing to fabricate sharp edged nanostructures composed of toxic metals (e.g. silver and iron oxide nanoparticles) to test the evolvability of resistance to poisoning, stabbing, or both on these surfaces. We will evaluate the physical characteristics of resistant clones and their genomic architecture should such emerge, as well as test the sustainability of these nano surfaces as antimicrobials.

BEACON@A&T Infrastructure Request

Personnel: Gerry Dozier, Joseph L Graves, Future Staff
Keywords: Administration
Abstract: Funds are requested for BEACON@A&T Infrastructure.

EXPLORING NEW APPROACHES TO ENSEMBLE AND NEURAL NETWORK TOPOLOGICAL EVOLUTION FOR TOUCH-BASED AUTHENTICATION

Personnel: NCA&T Graduate Student, Sandeep Kulkarni, Kaushik Roy
Keywords: Biometrics, Machine Learning
Abstract: In a typical access-control scenario, once authentication has taken place, a session remains active and a device remains unlocked until an individual or the server closes the session. In the event that a user leaves their device or workstation without closing an active session, the system may be accessed by an unauthorized party during the period before session timeout. To counteract this, active authentication can be implemented to continuously monitor the identity of the individual using a device. Behavioral biometrics can be monitored unobtrusively in the background, without the need to explicitly query the user for input. In this work, we will use touch-screen swipe pattern authentication to avoid smart-phone session hijacking.
This work will proceed in three phases. First is to collect data and establish a set of baseline classifiers for performance comparison. These will include a traditional feature-distance-based classifier, a Convolutional Neural Network (CNN), and a Recurrent Neural Network (RNN). Phase two will convert the feature-based classifiers from phase 1 into neural networks, and use them as a seed population to evolve an optimum ensemble using diversity as part of the fitness function. We will then apply stacked-generalization in an effort to further improve ensemble accuracy. Phase three will explore the use of a learned fitness function to accelerate the fitness evaluation in the evolution of domain-specific deep-nets.

Resistance is “fertile”: Antimicrobial resistance and the pleiotropic effects of compensatory mutations

Personnel: Peter Conlin, Jude Akamu Ewunkem, Joseph L Graves, Scott H. Harrison, Ben Kerr, Kristen Rhinehardt, Misty Dawn Thomas
Keywords: Antibiotic Resistance, Compensatory Evolution, Epistasis
Abstract: Mutations conferring resistance to antimicrobials commonly impose a fitness cost in the absence of selecting drugs. Removal of the selecting agent should therefore favor reversion back to sensitivity. Contrary to this expectation, experimental studies have found that resistance is often stably maintained in populations of drug resistant microorganisms even in the absence of antibiotics. This occurs when bacteria acquire second-site mutations that ameliorate the cost of resistance, called compensatory mutations. While many studies have demonstrated the beneficial fitness effects of compensatory mutations, few have thoroughly examined their effects on drug resistance (beyond providing confirmation that bacteria have not reverted back to sensitivity). Here we suggest that compensation may also lead to changes in the level of resistance due to epistatic effects with the initial mutation conferring resistance. We hypothesize that such epistatic interactions may be particularly common in cases where compensatory mutations occur within the same gene as the initial resistance mutation. Here, we propose to test this hypothesis using a combination of experimental evolution, genetic engineering, and whole genome sequencing to compare compensatory mutations in diverse mutant genotypes of Escherichia coli resistant to two distinct antimicrobial compounds: rifampicin and ionic silver (Ag+).

Proactively Evolving Communication Architectures of Malicious Android Apps

Personnel: Mohd Anwar, Gerry Dozier, PhD NCAT, Future Other
Keywords: Evolutionary Algorithm, Genetic Algorithm, Malware Detection, Neural Networks
Abstract: The sheer number and diversification of Android malware continues to increase at a drastic rate. This budget request proposes a research plan with the goal of evolving communication architectures to proactively identify Android malware. The proposed research simulates a study of evolution in the design of Android malware. The project will offer REU summer program for students at Guildford Technical Community College. The software developed through this project will be presented to the North Carolina Department of Information Technology for potential use.

UI Administration

Personnel: James Arthur Foster, Terence Soule, Future Staff
Keywords:
Abstract: We will begin scaling back this administrative request this year, consistent with the overall tapering of BEACON funding. Changes from previous requests are in italics.
We are requesting operating funds for Year 8, with which to accomplish the following goals:
a) Increase IBEST/BEACON participation and respond to unexpected opportunities;
b) Support BEACON staffing;
c) Transition to a sustainable post-NSFSTC environment

Microbes and Microplastics: Is there evidence for selection in a novel deep ocean niche?

Personnel: BEACON funded undergraduate researcher, James Arthur Foster, NCA&T Graduate Student, Joseph L Graves, Scott H. Harrison, Eric Mittelstaedt, Janet Williams
Keywords: Bacterial Populations, Biological Evolution
Abstract: Plastics are a serious threat to ocean ecosystems. Plastics are bio- and photo-degraded into tiny fragments that are being found in amounts on the deep-sea sediments of the ocean floor comparable to levels found in sediments of shallow and intertidal sea zones. Recent studies have identified plastics to be almost ubiquitous across samples assayed in several deep-sea environments including the North Atlantic, the Southern Ocean, the Gulf of Guinea, and the Mediterranean Sea (Van Cauwenberghe et al., 2013; Woodall et al., 2014). Another study has examined how this “plastic rain” effects the ocean floor ecosystem, finding that six of nine animals that were examined from diverse parts of earth’s oceans, have microfiber plastics in their systems (Taylor et al., 2016). It is reasonable therefore to suppose that there are recent (in evolutionary terms) emergence of niches for plastic-degrading bacteria (e.g., Zettler et al., 2013) in deep ocean sediment that would lead to selection for those bacteria. One of us (Mittelstaedt) recently collected ocean sediment cores from the Mid-Atlantic ridge (13°N -14°N). We propose to quantify changes in microbial communities’ composition and strength of selection across sediment locations and layers, and to examine plastic fragments directly for bacterial inhabitants. In short, we study evolution in action, by comparing pre-plastic and no plastic sediment layers and locations respectively, to how other parts of the deep ocean sediment microbiome have been impacted by the presence of plastic. This occurs in conjunction with our study of a unique and largely unexplored plastic cycle from the Mid-Atlantic surface to the sea floor.

BEACON TEACHERs Researching EVOLUTION (TEACHER REVOLUTION)_National

Personnel: Katie Dickinson, Kara Haas, Louise Souther Mead, Christine Parent, Tessa Solomon-Lane, Future Undergraduate Student, Jack Sullivan, Luis Zaman
Keywords: Informal Education, K-12 Education, Professional Development, Public Outreach
Abstract: The National BEACON TEACHER REVOLUTION Program (TEACHERs Researching EVOLUTION) will provide current K-12 teachers with a summer research experience in a mentored lab while simultaneously offering a professional development experience that identifies effective ways to integrate their research experience and science practices into their classroom. Teachers will have the opportunity to carry out original research, under the guidance scientists mentors at our consortium institutions. Specifically, we propose to place two teachers each at University of Washington, University of Idaho, University of Texas, and Michigan State University. A networked learning community will be established via weekly professional development via videoconferencing. This program builds on successful, evidence-based, programs that show research experiences positively impact the teachers as well as their students. Teachers will expand their understanding of evolutionary biology and the nature of science. They will impact their research lab by offering new perspectives and creating lesson plans and web-based materials to share their experience with a broad audience. BEACON’s legacy of exemplary outreach with K-12 teachers will continue and expand with this successful, national Research Experience for Teachers program

It only gets worse from here: adaptation in the face of ever-larger pulses of stress

Personnel: Jannell Bazurto, Emily Louise Dolson, Ben Kerr, Christopher J Marx, Sonia Singhal
Keywords: Adaptation
Abstract: Lowering rates of environmental change can expand possible evolutionary trajectories (Lindsey et al. 2013). We propose to evolve Methylobacterium extorquens for formaldehyde growth using varying rates of change in formaldehyde concentration, and to use an NK model to characterize evolution on fitness landscapes that differ between growth and survivability. In the Marx lab, we previously evolved M. extorquens PA1 experimentally to grow on formaldehyde using step-wise increments of near-lethal doses of formaldehyde. The resulting populations showed that growth on formaldehyde could be achieve by single mutations that activated uncharacterized stress response systems and/or rescued M. extorquens from extinction. Here, we propose to use selective regimes that go through sub inhibitory and sub lethal incremental steps to enrich for more paths toward formaldehyde growth. Lower rates of change may also provide an opportunity for improved formaldehyde utilization (intrinsically linked to detoxification) to be a significant selective pressure, which may be in direct competition with survivability. To inform and extend the results from M. extorquens, the Kerr and Ofria labs will implement an NK model that incorporates varying rates of environmental change and imposes different selective pressures during survival and growth. This combination of empirical and computational work will inform our understanding of how and why organisms adapt differently depending on the rate of environmental change.

The “staying power” of plasmids: The evolution of plasmid persistence and host permissiveness and their effects on the emergence of multi-drug resistance

Personnel: Hannah Jordt, Ben Kerr, Thibault Stalder, Future Undergraduate Student, Future Undergraduate Student, Eva M Top
Keywords: Antibiotic Resistance, Coevolution, Experimental Evolution, Mathematical Modeling, Plasmid
Abstract: Antibiotic resistance is commonly spread through bacterial populations via conjugative plasmids, extra-chromosomal pieces of DNA that can transfer between bacteria and often contain antibiotic resistance genes. Plasmids are generally considered costly towards their host in the absence of selection for the plasmid (e.g. an antibiotic-free environment). However, coevolution between a host and its plasmid can reduce plasmid carriage costs through compensatory mutations. While this can lead to greater maintenance of that plasmid in the population (and thus an increased proportion of antibiotic-resistant cells), it is unclear how these mutations affect the persistence of other novel plasmids in the environment, i.e. plasmid permissiveness. In aim 1 of this study, we will test the effects of host/plasmid coevolution on plasmid persistence and permissiveness by evolving multiple host/plasmid pairs and comparing the ability of the evolved vs ancestral hosts to maintain their own and novel plasmids in the absence of antibiotic selection. In aim 2, we will co-culture compensated pairs obtained from aim 1 and determine whether there is an increase in multi-drug resistance (resulting from the co-occurrence of both plasmid types in the same host), compared to when a co-culture of the ancestral pairs is propagated. We will further test whether this can lead to the creation of multi-drug resistant plasmids via recombination of both plasmid types into a single larger plasmid. These experiments will be complemented with mathematical modeling. Through this study we will gain a better understanding of how multi-drug resistant plasmids and bacterial populations arise, which has important relevance in our fight to mitigate multi-drug resistance in agricultural and clinical settings.

Off the deep end of the gene pool: Deep-sequencing and metabolomics analysis as a framework for uncovering fitness
landscapes and modeling evolvability in engineered metabolic networks

Personnel: Eric Bruger, Student Graduate, Cecilia Martinez, Christopher J Marx, Sergey Stolyar
Keywords: Adaptation, Bacterial Populations, Biotechnology, Epistasis, Evolvability, Fitness Landscapes, Gene Networks, Mathematical Modeling, Synthetic Biology, systems biology
Abstract: Evolution in many systems is fraught with constraints that limit the paths that are adaptive. Many combinations of genetic changes result in nonlinear fitness effects, or epistasis. Epistasis can potentially reduce (magnitude epistasis) or in extreme cases reverse (sign epistasis) the combined fitness between alleles. From the perspective of an adaptive landscape, epistasis adds ruggedness, making it more difficult for populations to traverse the landscape to effectively search and find fitness peaks. This problem presents itself in an system with sufficient epistasis, including both naturally evolving populations as well as attempts to optimally engineer organisms for industrial applications. Epistasis can be identified at many levels of organization, and is very likely to occur within metabolic pathways. We propose to study epistasis at an unprecedented level of resolution for a metabolic network by adopting a new method to generate large degrees of genotypic variation on short time scales and assessing changes in pooled populations through deep-sequencing. Combined short- and long-read sequencing results will provide large amounts of genotype-fitness data, which we will choose among to pursue more in-depth phenotypic characterization by measuring enzyme and metabolite levels, and fitting mathematical models of the entire network. The result of this combined approach will be a better understanding of epistasis in this metabolic system and an overall approach to identifying and addressing epistatic interactions broadly.

Scientific Virtues Toolbox: Meeting RCR Needs for BEACON – And Beyond!

Personnel: Eric Berling, Robert Heckendorn, Michael O’Rourke, Robert T. Pennock
Keywords: Responsible Conduct Of Research
Abstract: The Scientific Virtues Toolbox will develop responsible conduct of research modules and conduct workshops that will be available for all BEACON participants for Friday seminar and Congress. It will collect pilot data to be used for external grant opportunities that we have already identified.

SECOND YEAR: Rapid analysis of parasite dynamics and evolution in arthropod populations

Personnel: Andrew D Ellington, Postdoctoral Researcher, LuAnn Scott, Unnamed Undergraduate Student, Multiple Undergraduate Students, Holly Wichman
Keywords: Host-Parasite Interactions, Medical Diagnostics, Population Genomics, Synthetic Biology, Virus
Abstract: Many viruses that afflict and threaten humanity are carried by insects, such as the recent Zika virus outbreak that is carried by mosquitoes. Being able to monitor the spread of the virus through insect and human populations, and being able to chart viral evolution, are of key importance for public health decisions. However, it is difficult to monitor viruses in the wild, because we lack instrumentation to do so. It is also difficult to develop instrumentation, because experiments with diseases require extensive and expensive containment. We have solved these problems by combining the expertise of the Ellington lab at Texas and the Wichman lab at Idaho. The Ellington lab has developed a portable device that can ascertain molecular signatures of infection in the field. The Wichman lab has developed a model system for infecting the laboratory organism Drosophila with arboviruses. We propose to use the point-of-care device to monitor the spread of arboviruses in Drosophila, with an eye towards understanding whether this exercise could be extended to mosquito populations and more deadly arboviruses in the wild. In addition, the Drosophila system can be used to monitor the spread of the bacteria Wolbachia, which is being considered as a countermeasure to arboviruses. However, there are legitimate concerns about containing countermeasures, and a study that can measure interactions between host (Drosophila), pathogen (arbovirus), and countermeasure (Wolbachia) will greatly inform the scientific and public health communities regarding how to best intervene in insect-borne epidemics.

Computational studies to elucidate evolutionary conservation of phosphorylation sites

Personnel: NCA&T Graduate Student, Student Graduate, Dukka KC, Robert Newman, Claus O. Wilke
Keywords: Computational Biology
Abstract: Phosphorylation is a key posttranslational modification that controls the regulation of nearly all cellular processes. Importantly, because cellular phosphorylation networks play a central role in regulating gene expression patterns and cellular metabolism, re-organization of these networks is likely to lead to systems-level changes in cellular physiology that can impact evolutionary parameters in a manner analogous (or complementary) to genetic variation. Interestingly, though phosphorylation-dependent signaling is a highly conserved mode of information processing across eukaryotic species, recent studies suggest that individual phosphorylation sites on orthologous proteins can vary substantially across species. While many studies have focused on individual phosphorylation sites, there are relatively few that have explored evolutionary conservation of phosphosites across various species.
To perform a systematic comparison of phosphorylation network between species, we first must expand phosphosite datasets for various organisms. We must then develop an approach to compare phosphorylation sites across species. Thus, in this pilot project, we will pursue two primary objectives: First, we will build on our previous studies to develop a generalizable approach for the development of phosphorylation site prediction tools that utilize machine learning approaches. This will ultimately allow us to enrich the information about phosphosites in various organisms. Second, we will develop an approach to perform comparative analyses of phosphorylation sites across species. These algorithms & datasets will be useful for exploring the evolutionary conservation of phosphorylation sites across species and for understanding the rules underlying phosphorylation-dependent signaling.

BEACON and Beyond: Broadening Participation and the Scope of Public

Personnel: Travis Hagey, Hans A. Hofmann, Tessa Solomon-Lane, Alexa Warwick
Keywords: Graduate Education, K-12 Education, Postdoc Education, Public Outreach, Science Education
Abstract: Engaging with the public is a fundamental professional responsibility for academic scientists, and it is required by many funding agencies. Engagement is also a civic responsibility. Although Americans have confidence in science and trust in scientists, science denial and the spread of false claims, especially in politics, can be dangerous for society. However, participation in public engagement is often limited to “the dedicated few,” as many are unaware that public engagement benefits professional development and scholarship. We have developed an innovative, evidence-based program that motivates, recruits, and trains graduate students and postdocs in best practices for public engagement and pairs them with opportunities to engage. We aim to professionally package and disseminate this program so it can be replicated in any institution to concurrently enhance trainee professional development and public engagement. We will also expand our program to include a module on framing a scientific audience for different audiences, one of the most important communication skills. Finally, although BEACONites currently engage the public in a wide variety of ways, our survey data show striking underrepresentation of engagement with public policy and policy makers. We will develop a specialized curriculum for engaging this audience and facilitate policy engagement opportunities. This work adds new and important dimensions to public engagement training and action and will reshape how scientists learn to communicate to diverse and influential audiences.

Decoy mutations: engineering fitness landscapes to delay unwanted evolution

Personnel: Christoph Adami, Jeffrey E Barrick, Future Graduate Student, Thomas LaBar, Luis Zaman
Keywords:
Abstract: Synthetic biology is rapidly advancing our ability to create complex genetic circuits and perform feats of metabolic engineering. This progress is enabling medical, manufacturing, and energy breakthroughs. However, when organisms are engineered to perform some desired task, it is typically at the expense of resources that would otherwise be invested in producing more offspring. For this reason, evolution often favors organisms that have accumulated mutations disrupting the engineered function. Here, we propose to harness clonal interference, a phenomenon that occurs in large asexual populations whereby competition between multiple beneficial mutations slows the fixation of any one allele, as a means of prolonging the lifetime of a burdensome function in a population. The basic setup is to add the possibility for one or more decoy mutations into the fitness landscape of a cell. The decoy mutations are designed to “distract” adaptive evolution from inactivating the engineered function. Our two key questions for this proof-of-principle seed project are: (1) What is the optimal design for a suite of decoy mutations to maximally extend the lifetime of the desired function in a population? (2) Can we extend the half-life of an engineered trait even further by dynamically altering the environment in a way that replenishes the supply of decoy mutations? We will examine both questions using a BEACON-inspired combination of modeling approaches and microbial evolution experiments.

The University of Texas Infrastructure Request

Personnel: Laurie Alvarez, Risto Miikkulainen
Keywords:
Abstract: Funds are requested for the infrastructure supporting BEACON projects at the University of Texas.

Experimental evolution of microbiomes for ‘studs’ and ‘duds’ in Drosophila melanogaster

Personnel: Future Graduate Student, Chad Christopher Smith, Danielle J. Whittaker
Keywords: Communication, Evolution In Action, Experimental Evolution, Mate Choice, Metagenomics, Mutualism, Sexual Selection, animal behavior, bacteria
Abstract: Animals are host to diverse communities of microbial symbionts that have played an integral role in their ecology, evolution, and behavior. Archie & Theis identified sexual selection as one of three most promising areas at the interface between animal behavior and microbial ecology, for at least two reasons. First, sexually selected traits such as ornaments and weapons are highly condition-dependent, and the microbiome is central to nutrition and immune function during development. Second, microbes influence animal communication by contributing to the production of signals by metabolizing host substrates, particularly in olfactory communication. We propose to test whether ecological and evolutionary changes in microbiomes affect male sexual attractiveness in Drosophila melanogaster using ‘one-sided’ host-mediated selection on microbiomes. We will produce ‘stud’, ‘dud’, and ‘random control’ microbiome selection lines by propagating the microbiomes of attractive, unattractive males, and random males over five selection cycles against a static host genetic background. Illumina sequencing of 16S rRNA amplicons will be used to determine whether selection lines differ in microbiome structure and to understand how changes in community states vary over the course of the experiment. To assess the proximate cause of variation in attractiveness between treatments, courtship rate and cuticular hydrocarbons will be measured to determine whether microbiomes affect sexual behavior and olfactory signals important in sexual selection, which has previously been suggested in D. melanogaster. We will also measure several life history variables to determine whether microbiomes influence covariation with sexually selected traits.

SECOND YEAR: Modeling the Evolution and De-Evolution of Biological Complexity

Personnel: Christoph Adami, Thomas LaBar, Richard Lenski, Dariya K. Sydykova, Claus O. Wilke
Keywords: Adaptive Complexity, Digital Evolution, Experimental Evolution, Mathematical Modeling
Abstract: The mathematical description of evolving living systems has made great strides in the last century, but theory is concerned mostly with understanding evolution on short time scales: those relevant for the fixation of a single or perhaps a pair of loci. Modern developments in experimental evolution have highlighted a missing topic in the mathematics of population genetics, namely a theory that describes how populations adapt over the long term. A recent theory by Wiser Ribeck, and Lenski (WRL theory) has shown that evolution over the long term can be predicted given a sufficiently long trajectory in the past. It turns out that any theory that can predict forward evolution, should also be able to predict the loss of fitness over time, due to drift (for example when population sizes are sma
ll. We can use such a theoretical prediction to estimate the complexity of the organism that is losing fitness, as several theoretical approaches suggest that the rate of fitness loss is determined by the complexity of the initial organism. We will develop the theory to estimate the fitness loss over long-term, and validate it by performing drift-response experiments with well-characterized genomes of the Long-Term Evolution Experiment (LTEE). If successful, we will have a new quantitative way to link fitness to organismal complexity.

SECOND YEAR: Rapid analysis of parasite dynamics and evolution in arthropod populations

Personnel: Andrew D Ellington, Postdoctoral Researcher, LuAnn Scott, Unnamed Undergraduate Student, Multiple Undergraduate Students, Holly Wichman
Keywords: Host-Parasite Interactions, Medical Diagnostics, Population Genomics, Synthetic Biology, Virus
Abstract: Many viruses that afflict and threaten humanity are carried by insects, such as the recent Zika virus outbreak that is carried by mosquitoes. Being able to monitor the spread of the virus through insect and human populations, and being able to chart viral evolution, are of key importance for public health decisions. However, it is difficult to monitor viruses in the wild, because we lack instrumentation to do so. It is also difficult to develop instrumentation, because experiments with diseases require extensive and expensive containment. We have solved these problems by combining the expertise of the Ellington lab at Texas and the Wichman lab at Idaho. The Ellington lab has developed a portable device that can ascertain molecular signatures of infection in the field. The Wichman lab has developed a model system for infecting the laboratory organism Drosophila with arboviruses. We propose to use the point-of-care device to monitor the spread of arboviruses in Drosophila, with an eye towards understanding whether this exercise could be extended to mosquito populations and more deadly arboviruses in the wild. In addition, the Drosophila system can be used to monitor the spread of the bacteria Wolbachia, which is being considered as a countermeasure to arboviruses. However, there are legitimate concerns about containing countermeasures, and a study that can measure interactions between host (Drosophila), pathogen (arbovirus), and countermeasure (Wolbachia) will greatly inform the scientific and public health communities regarding how to best intervene in insect-borne epidemics.

Tradeoffs During Niche Adaptation Stabilize Niche Construction

Personnel: Brian Connelly, Ben Kerr, Joshua Richard Nahum, Undrgraduate Student, Katrina van Raay
Keywords: Agent Based Modeling , Cooperation, Eco-evolutionary processes, niche construction, tradeoff
Abstract: Through just about all that they do, organisms change their environment. From metabolizing resources to building dams, these niche constructing behaviors can profoundly influence natural selection in both positive and negative ways. This project will develop and test a new theory that as populations modify their niche, tradeoffs can emerge that enable positive niche constructing behaviors to be maintained by a paper-rock-scissors dynamic. We will also extend our model to include epistasis, which may lead populations to become “addicted” to niche construction as they continue to adapt to their environment.

Decoy mutations: engineering fitness landscapes to delay unwanted evolution

Personnel: Christoph Adami, Jeffrey E Barrick, Future Graduate Student, Thomas LaBar, Luis Zaman
Keywords:
Abstract: Synthetic biology is rapidly advancing our ability to create complex genetic circuits and perform feats of metabolic engineering. This progress is enabling medical, manufacturing, and energy breakthroughs. However, when organisms are engineered to perform some desired task, it is typically at the expense of resources that would otherwise be invested in producing more offspring. For this reason, evolution often favors organisms that have accumulated mutations disrupting the engineered function. Here, we propose to harness clonal interference, a phenomenon that occurs in large asexual populations whereby competition between multiple beneficial mutations slows the fixation of any one allele, as a means of prolonging the lifetime of a burdensome function in a population. The basic setup is to add the possibility for one or more decoy mutations into the fitness landscape of a cell. The decoy mutations are designed to “distract” adaptive evolution from inactivating the engineered function. Our two key questions for this proof-of-principle seed project are: (1) What is the optimal design for a suite of decoy mutations to maximally extend the lifetime of the desired function in a population? (2) Can we extend the half-life of an engineered trait even further by dynamically altering the environment in a way that replenishes the supply of decoy mutations? We will examine both questions using a BEACON-inspired combination of modeling approaches and microbial evolution experiments.

Plasticity and epigenetic modification as a form of bet-hedging in a highly variable environment.

Personnel: BEACON funded undergraduate researcher, Cynthia Chang
Keywords: Climate Change
Abstract: Climate models predict a more variable environment with longer periods of drought followed by intense rainfall. It is important to understand how plant populations will respond and adapt to these predicted environmental changes. Environmentally induced epigenetic modifications could provide the basis for phenotypic plasticity and adaptive ability to drought and variable precipitation regimes, where the intensity and length of stress is unpredictable. Epigenetic “memory” of a high stress experience , could be potentially beneficial to the offspring in a highly variable stress environment. Despite the importance of understanding the role of epigenetics in evolution and ecology, the adaptive significance of heritable epigenetic modification and its relationship to phenotypic plasticity still remains unclear. To our knowledge, no study has tested the importance of epigenetic memory from a high stress environment on a progeny’s ability to deal with a variable environment that incorporates both high and low stress at unpredictable intervals. Herein, we use the model plant Arabidopsis thaliana to 1) address the transgenerational epigenetic consequences of drought, 2) explore how epigenetic changes influence phenotypic plasticity, and 3) evaluate the advantage of epigenetics as a form of evolutionary bet-hedging in a highly variable environment. As part of the broader impacts, the project harnesses an existing research program by conducting course-based research with undergraduates at the University of Washington Bothell (a primarily undergraduate institution).

It only gets worse from here: adaptation in the face of ever-larger pulses of stress

Personnel: Jannell Bazurto, Emily Louise Dolson, Ben Kerr, Christopher J Marx, Sonia Singhal
Keywords: Adaptation
Abstract: Lowering rates of environmental change can expand possible evolutionary trajectories (Lindsey et al. 2013). We propose to evolve Methylobacterium extorquens for formaldehyde growth using varying rates of change in formaldehyde concentration, and to use an NK model to characterize evolution on fitness landscapes that differ between growth and survivability. In the Marx lab, we previously evolved M. extorquens PA1 experimentally to grow on formaldehyde using step-wise increments of near-lethal doses of formaldehyde. The resulting populations showed that growth on formaldehyde could be achieve by single mutations that activated uncharacterized stress response systems and/or
rescued M. extorquens from extinction. Here, we propose to use selective regimes that go through sub inhibitory and sub lethal incremental steps to enrich for more paths toward formaldehyde growth. Lower rates of change may also provide an opportunity for improved formaldehyde utilization (intrinsically linked to detoxification) to be a significant selective pressure, which may be in direct competition with survivability. To inform and extend the results from M. extorquens, the Kerr and Ofria labs will implement an NK model that incorporates varying rates of environmental change and imposes different selective pressures during survival and growth. This combination of empirical and computational work will inform our understanding of how and why organisms adapt differently depending on the rate of environmental change.

The “staying power” of plasmids: The evolution of plasmid persistence and host permissiveness and their effects on the emergence of multi-drug resistance

Personnel: Hannah Jordt, Ben Kerr, Thibault Stalder, Future Undergraduate Student, Future Undergraduate Student, Eva M Top
Keywords: Antibiotic Resistance, Coevolution, Experimental Evolution, Mathematical Modeling, Plasmid
Abstract: Antibiotic resistance is commonly spread through bacterial populations via conjugative plasmids, extra-chromosomal pieces of DNA that can transfer between bacteria and often contain antibiotic resistance genes. Plasmids are generally considered costly towards their host in the absence of selection for the plasmid (e.g. an antibiotic-free environment). However, coevolution between a host and its plasmid can reduce plasmid carriage costs through compensatory mutations. While this can lead to greater maintenance of that plasmid in the population (and thus an increased proportion of antibiotic-resistant cells), it is unclear how these mutations affect the persistence of other novel plasmids in the environment, i.e. plasmid permissiveness. In aim 1 of this study, we will test the effects of host/plasmid coevolution on plasmid persistence and permissiveness by evolving multiple host/plasmid pairs and comparing the ability of the evolved vs ancestral hosts to maintain their own and novel plasmids in the absence of antibiotic selection. In aim 2, we will co-culture compensated pairs obtained from aim 1 and determine whether there is an increase in multi-drug resistance (resulting from the co-occurrence of both plasmid types in the same host), compared to when a co-culture of the ancestral pairs is propagated. We will further test whether this can lead to the creation of multi-drug resistant plasmids via recombination of both plasmid types into a single larger plasmid. These experiments will be complemented with mathematical modeling. Through this study we will gain a better understanding of how multi-drug resistant plasmids and bacterial populations arise, which has important relevance in our fight to mitigate multi-drug resistance in agricultural and clinical settings.

Resistance is “fertile”: Antimicrobial resistance and the pleiotropic effects of compensatory mutations

Personnel: Peter Conlin, Jude Akamu Ewunkem, Joseph L Graves, Scott H. Harrison, Ben Kerr, Kristen Rhinehardt, Misty Dawn Thomas
Keywords: Antibiotic Resistance, Compensatory Evolution, Epistasis
Abstract: Mutations conferring resistance to antimicrobials commonly impose a fitness cost in the absence of selecting drugs. Removal of the selecting agent should therefore favor reversion back to sensitivity. Contrary to this expectation, experimental studies have found that resistance is often stably maintained in populations of drug resistant microorganisms even in the absence of antibiotics. This occurs when bacteria acquire second-site mutations that ameliorate the cost of resistance, called compensatory mutations. While many studies have demonstrated the beneficial fitness effects of compensatory mutations, few have thoroughly examined their effects on drug resistance (beyond providing confirmation that bacteria have not reverted back to sensitivity). Here we suggest that compensation may also lead to changes in the level of resistance due to epistatic effects with the initial mutation conferring resistance. We hypothesize that such epistatic interactions may be particularly common in cases where compensatory mutations occur within the same gene as the initial resistance mutation. Here, we propose to test this hypothesis using a combination of experimental evolution, genetic engineering, and whole genome sequencing to compare compensatory mutations in diverse mutant genotypes of Escherichia coli resistant to two distinct antimicrobial compounds: rifampicin and ionic silver (Ag+).

UW Infrastructure 2017-2018

Personnel: Katie Dickinson, Ben Kerr
Keywords:
Abstract: (no abstract)

A one-two punch: combining killing systems within a single cell by going “pro”, and the mystery of the one-punch

Personnel: Joshua Richard Nahum, Future Undergraduate Student, Katrina van Raay
Keywords: Agent Based Modeling , Eco-evolutionary processes, Experimental Evolution
Abstract: Chemical warfare in the microbial world is ubiquitous. One kind of chemical weapon microbes employ is a proteinaceous toxin called a bacteriocin. The best-studied bacteriocins are the colicins, produced by and active against Escherichia coli. Many colicin systems encode suicidal lysis genes, such that the producing cell releases the toxin through cell lysis. Released colicin kills sensitive competing cells allowing immune clones of the producer to capitalize on the liberated resources. However a major group of colicin systems lack this lysis gene, and it has been unclear how such toxins were released from the producing cell. Here, we explore a newly discovered union between such a colicin system and a prophage (bacterial virus that has been incorporated into the bacterium’s genome), where release of the colicin occurs via phage-encoded cell lysis. This union should be detrimental to susceptible cells, which would be hit by a one-two punch: once by the colicin, once by the phage. Preliminary mathematical modeling suggests that the success of this dual killing system depends on the prophage’s ability to produce infectious virions. If the prophage is cryptic (encoding lysis, but not producing infectious phage), then this dual system is susceptible to invasion by “cheaters” that only possess the colicin system (and its immunity) but do not lyse. Here, we explore this social dilemma and its potential resolution in detail through a combination of mathematical modeling, microbial evolution experiments, and agent-based simulations.