symposium banner

Please feel free to use this banner to link to the
official website of the meeting here

 

 

Submissions Closed

Flyer


Download
Workshop Program
PDF Here
 

Abstracts of the Workshop

Thur-Fri, August 4-5th, 2016

Go back to Main Page

Last updated on 3rd August 2016

 

Welcome to EvoSysBio, Workshop Overview and Context:
Why Modeling Needs More than One Perspective

Dr. Laurence Loewe

Evolutionary Systems Biology Group, Wisconsin Institute for Discovery, UW-Madison

After a brief overview of the workshop, it will be argued that an efficient method for unmasking blind spots in modeling is to change perspectives or to invite other disciplines to contribute theirs. EvoSysBio will be more efficient as a research paradigm, if it can understand transdisciplinary integration as a union that inspires further investigation and not as an intersection of (usually unattainable) many areas.

.

 

Invited Speaker: Dr. Mentewab Ayalew
Department of Biology,
Spelman College, Atlanta, GA, USA

Keynote: Are antibiotic resistance mechanisms different in plants and bacteria?

Plants are exposed to antibiotics produced by soil microorganisms, but little is known about mechanisms of resistance. The Arabidopsis thaliana ATP Binding Cassette (ABC) transporter AtWBC19 (ABCG19) is one of two genes known to affect antibiotic resistance in plants. ABC transporters are a family of proteins that typically use a common scaffold to move diverse ligands across membranes. They are found in all organisms, including bacteria where specific ABC proteins are also associated with antibiotic resistance. Although Atwbc19 is predicted to be a membrane protein, our preliminary results suggested that it is cleaved and may not function as a transporter. In order to propose an alternative mechanism of function, we used a number of bioinformatics tools and sought to identify proteins interacting with Atwbc19. Yeast two-hybrid screens were implemented and two proteins of interest - a subunit of initiation factor 3 and a kinase were identified. Further characterization of Atwbc19 established that it is mainly expressed in the vascular tissues of plants and thus may affect the long distance transport of kanamycin from the soil to the leaves. In addition, Atwbc19 affected the zinc and iron contents of plants. Our findings suggest that a novel mechanism of resistance that involves translational regulation might be associated with Atwbc19.

 

Dr. Ye Xu
with Kalin Vetsigian

Department of Bacteriology and Wisconsin Institute for Discovery, UW-Madison

Germination Interactions among Streptomyces

Bacteria live in crowded and diverse communities and exhibit a wide range of social behaviors mediated by diffusing chemicals. While many types of interactions have been exhaustively studied, little is known about how the germination decisions of individual spores are affected by the activities of their microbial neighbors. In particular, it is currently unknown how widespread or diverse microbial germination interactions are, and there has been no systematic investigation of the spatial and temporal correlations of germinations they cause in communities of spores. Focusing on antibiotic producing bacteria, Streptomyces, we developed a high-throughput platform for automated quantification of spore germination on 2D surfaces through time-lapse microscopy. Fluorescent labeling allows us to investigate single- and two-species communities. By monitoring how the germination probability as a function of time is affected by the spore density and the germination of neighboring spores, we quantified the germination interactions among a panel of Streptomyces strains. All species exhibited distinct germination interactions. While germlings from some species promoted the germination of their sister spores, other species inhibited their own germination and still others exhibited no interactions. We also found strong positive and negative inter-species interactions that are mediated by defusing chemicals. Furthermore, we discovered that as spores of certain species age they enter a state of deep dormancy that can be fully reversed by chemicals released during germination. This work reveals that germination interactions among Streptomyces are common and diverse. Thus, bacterial germination is commonly affected by the community context, and different species have adapted different germination strategies.

 

Erik Wright
with Kalin Vetsigian
Department of Bacteriology and Wisconsin Institute for Discovery, UW-Madison

Survival of the Common is Nearly as Widespread as Survival of the Fittest among Antibiotic Producing Bacteria

Knowledge of the patterns of ecological interactions and their frequency dependence within and between species is central to understanding ecological and evolutionary dynamics in communities. At the same time, the type and pattern of microbial interactions that exist in natural communities are poorly understood, which is limiting our comprehension of ecological systems and our ability to engineer microbial consortia. Here we determined the network of invasions and inhibitions among a panel of 18 naturally antibiotic producing bacteria. We performed a tournament between all species pairs beginning from two extremes of relative abundance, and determined whether the least abundant strain could invade using high-throughput sequencing. In addition to survival of the fittest, we observed many cases of 'survival of the common' where the most abundant strain is the winner. Surprisingly, we observed few cases of cycles of dominance, and no instances of the well-studied rock-paper-scissors loop. The invasion network was shaped by inhibitory interactions between strains, which also promoted survival of the common. These findings have several immediate implications for how we view the assembly, structuring, and diversity of microbial communities. They indicate that pairwise interactions exhibit inherent nonlinearities that predispose communities toward multiple stable states. This makes microbial communities intrinsically sensitive to initial conditions during community assembly but, at the same time, makes them more resistant to change once they are established. The frequent 'survival of the common' that we identified may also promote mosaic distributions with different microbial populations dominating different spatial patches or hosts despite similar abiotic conditions.

 

 

Dr. Sushmita Roy
Systems Biology, Wisconsin Institute for Discovery,
Department of Biostatistics and Medical Informatics,
UW-Madison, WI, USA

Computational Approaches for Studying Regulatory Networks: From Inference to Evolution

Transcriptional regulatory networks specify the upstream regulatory programs of genes and control which genes must be expressed when. A major challenge is to decipher these regulatory networks in dynamic processes such as development or during evolution. Transcriptional regulatory network mapping is difficult because gene expression is controlled at multiple levels including transcription factor binding, chromatin state and three-dimensional organization of the genome. This tutorial will cover recent computational advances to map regulatory networks in diverse biological processes and to study their dynamics on developmental and evolutionary lineages.

 

 

Kate Scheuer with Jerdon Dresel, Erik Nolan, Bret Hanlon, Laurence Loewe
Evolutionary Systems Biology Group, Wisconsin Institute for Discovery, UW-Madison

FlyClockbase: a 25-year Overview of Time Series of the Drosophila melanogaster Circadian Clock

Circadian rhythms are governed by specific biochemical pathways which produce clock-like patterns of cyclical protein expression. The circadian clock of the fruit fly (Drosophila melanogaster) is a complex regulatory gene network which receives light and other environmental input and produces cyclical changes in behavior and protein expression through interlocking positive and negative feedback loops. Deciphering the inner workings of this clock is challenging, and mathematical modeling can be a powerful tool to this end. While models are most realistic when they are based on experimental data, collecting and utilizing this data can be difficult. Consequently, many models aim to satisfy a set of general requirements (e.g. period length) rather than predicting specific experimental time series (measurements of the relative amounts of one or more given mRNA or proteins over time). To facilitate the use of experimental data in mechanistic modeling, we compiled FlyClockbase, a file-based database of over 400 time series of circadian mRNA and proteins in D. melanogaster published between 1990 and 2015. Because it contains time series from the previous 25 years, this database is a unique resource which can be used to improve understanding of the D. melanogaster circadian clock. FlyClockbase provides access to a large number of time series which can be used for more realistic parameter estimation or comparison of simulation results and experimental data. It also offers a platform for comparing experimentally observed variability within and between different types of circadian mRNA and protein components. Identifying sources of such variability opens new research avenues and may help to better define boundaries for sub-models or potential biological mechanisms within the clock. In this talk, we present FlyClockbase and demonstrate sample use cases.

 

Dr. Brian McLoone
Evolutionary Systems Biology Group, Wisconsin Institute for Discovery, UW-Madison

A Multilevel N-Player Snowdrift Game and the Evolution of Cooperation

Evolutionary game theory combines insights from game theory with questions of evolvability. Specifically, the question of stable strategies are important; sometimes it is possible to demonstrate expected evolutionary stability by showing that a certain type of game strategy can win if it exists. This work combines theoretical modeling insights from the so-called snow-drift game and its extension to many players with biological results from analyzing genomes and biofilm formation in order to understand the evolutionary dynamics of multi-level selection underlying biofilm formation and colonization. There is a substantial amount of work on N-player snowdrift games, much work on multi-level selection, and some work providing game theoretic perspectives on biofilm formation. However, relevant questions are rarely analyzed together in an effort to arrive at a more comprehensive perspective on the evolution of cooperation that is linked to observable data.

 

Dr. Laurence Loewe
Evolutionary Systems Biology Group, Wisconsin Institute for Discovery, UW-Madison

Informal Discussion of various Models and Modeling Questions

There are deliberate gaps in this schedule to allow participants to informally discuss all sorts of modeling questions. While coffee breaks and lunch are great for starting such discussions, they don't always provide enough time for more substantial progress or to wrap up some aspects. This time is dedicated to provide such opportunities as the need arises. Obviously, not all modeling problems can be solved in the time we have; however, if biologists, modellers, computer scientists, physicists, and many other disciplinary scientists present put their mind to look into a certain modeling question, then an interesting lively trans-disciplinary dialogue is almost guaranteed: we will either be surprised by useful pieces of the puzzle that arrive from completely unexpected directions, or we will be surprised at how difficult it can be to cross some (disciplinary) language barriers, or both and much much more.
If you have a particular modeling problem that you would like to have discussed, please bring this to the attention of the workshop organizers, who will select one or more modeling questions received for discussion with those who are interested to stay for a bit longer at the workshop.

 

 

 

 

Introductory Remarks on Common Ground between Transdisciplinary Education and Programming Languages
Dr. Laurence Loewe

Evolutionary Systems Biology Group, Wisconsin Institute for Discovery, UW-Madison

A brief overview of how these two fit form an unexpected and powerful partnership when reproducibility is expected as a key outcome.

 

Dr. Mentewab Ayalew
Department of Biology,
Spelman College, Atlanta, GA, USA

How Biology and Physics Departments learned to collaborate on teaching

The Biology and Physics department at Spelman started to implement an NSF funded innovative pedagogy to bridge connections between the two disciplines. In this trans-disciplinary educational project, students are introduced to a problem scenario or challenge in the Introductory Biology class. Although they are introduced to basic physics concepts, they build models to test their biological hypothesis; then analyze and present their data. As they move to their Introductory Physics course, they revisit the same problem, this time completely from the disciplinary perspective of physics. Finally a more integrated analysis of the problem can be done in their upper level Biology course. This approach enabled the integration of biology into physics courses and vice versa, allowing biology majors to see their curriculum as a unified whole. We plan to extend the use of this educational model to connect biology with other disciplines, especially math and computer science as well as our own upper level courses.

 

Dr. John Yin
Chemical Engineering, Wisconsin Institute for Discovery, UW-Madison
Measuring what counts: Trans-disciplinary aspects of an engineering education illustrated while quantifying mechanisms of virus growth.

The shortest programs of organismal development are encoded in the genomes of viruses, bits of genetic material that reproduce by directing the biosynthetic functions of living cells toward their own replication, spread, and persistence. This talk will share cross-disciplinary challenges and opportunities for quantifying and predicting virus growth and infection spread.

 

Dr. Laurence Loewe
Dr. Brian McLoone
Evolutionary Systems Biology Group, Wisconsin Institute for Discovery, UW-Madison

Experiences with a Trans-disciplinary EvoSysBio Course in an Efficient Disciplinary Education Environment

Why trans-disciplinary research can be easy and difficult at the same time. Experiences with a problem-based learning environment that exposes students to the reality of research.

 

Everybody
Evolutionary Systems Biology and Modeling Workshop 2016 at the Wisconsin Institute for Discovery, UW-Madison

An Open Discussion about Inspiring Trans-Disciplinary Work in EvoSysBio and Beyond

Share your experiences, bring your questions. Let's see how we can build on what we've learned about inspiring students and colleagues to explore trans-disciplinarity.

 

 

Everybody
Evolutionary Systems Biology and Modeling Workshop 2016 at the Wisconsin Institute for Discovery, UW-Madison

We will attempt to form groups around some modeling questions that we hope will be discussed at luch - at least for a little bit. Each group can decide by themselves which of Madison's many nice restaurants they want to choose for lunch.

Let's see how much biology can fit on napkins and how biology is represented there, often in a very intuitive way. The idea is, to learn form the efficiency that many biologists use in such informal settings to inspire efficient syntax design for Evolvix (some of it to be reviewed after Lunch). So, share your experiences, bring your questions.

 

 

Dr. Laurence Loewe
Evolutionary Systems Biology Group, Wisconsin Institute for Discovery, UW-Madison

The Importance of Names in Biological Modeling and the Flipped Programming Language Design Approach

Recent conceptual advances in how to handle synonyms and dialects of names that serve different use-cases have opened new possibilities for engaging a disciplinary target audience (such as biologists) in designing a programming language that serves their needs best. Some unexpected features of this development process will be discussed. This session serves as the foundation for the next two. The goal is to attempt to work together in reviewing proposed syntax for Evolvix by testing how well the design holds up against the notorious potential for human languages to introduce ambiguities into statements that require human-level intelligence to resolve appropriately. Asking users first has been an efficient method for facilitating the simplification of Evolvix.

 

Lea Rogers
Dr. Laurence Loewe
Spelman College, Atlanta, GA; Wisconsin Institute for Discovery, UW-Madison

Pain(t) by Logic: The Ambiguity Game

This game is entirely logical! Identify the areas of a Venn diagram selected by a given logic operator and color them as fast as you can. Then find all linguistically or otherwise justifiable ways of how you could mis-interpret the meaning of this particular choice of word for a logic operator. Given the crystal-clear sharp nature and simplicity of Binary Logic as developed by Bool and others in the last century, it should be simple to find unambiguous English words that can represent the 16 binary operators that are so foundational for many algorithms in computer science. We argue that this isn't simple and will ask the audience to particpate in a game designed to help us get to greater clarity in how to syntactically represent these operators that are important for many biological questions that could be efficiently solved with computers - if an intuitive bio-friendly way could be found for representing these concepts. There are prices to be won here, so put your BEST Names forward...

 

Dr. Laurence Loewe
Evolutionary Systems Biology Group, Wisconsin Institute for Discovery, UW-Madison

Discussion of Fundamental Modes of Computing: How can we integrate this and be user friendly?

All general purpose programming languages integrate some aspects of what could be termed "fundamental modes of computing". Recently, some of these (or maybe all?) have been (re)discovered in an attempt to integrate and simplify how these can be combined in Evolvix. See also the poster on this topic. We will review these and ask how their integration may or may not be simplified. The aim is to find a syntax that works (ideally) equally well for non-computing biologists and professional computer scientists. If this turns out to be impossible, then this session will try to get as close to that ideal as possible. One or more practical examples may be embedded in the discussions and may also be contributed by the audience.

 

Dr. Mentewab Ayalew
Department of Biology, Spelman College, Atlanta, GA, USA
Dr. Laurence Loewe
Evolutionary Systems Biology Group, Wisconsin Institute for Discovery, UW-Madison

Closing Remarks

Potential next steps. To be continued by an informal discusion.

 

 

 

 

Dr. Wai-Tong (Louis) Fan
with Amarjit Budhiraja, Zhenqing-Chen, Rick Durrett,
Mathematics, UW-Madison

Stochastic spatial models for complex systems

Mathematicians and scientists use interacting particle models to gain understanding of the emergence of macroscopic phenomena from microscopic laws of nature. In this poster, I will illustrate the use of these powerful tools by practical examples including the transport of charges in solar cells and the evolution of randomly growing cancer tumors. To connect the microscopic mechanisms with the macroscopic behaviors at two different scales, we obtain the functional law of large numbers for these systems. The limiting objects are described by partial differential equations and stochastic partial differential equations.

 

Dr. Laurence Loewe
Genetics, UW-Madison
Evolutionary Systems Biology: Overview and Challenges

Collection of some posters providing an overview and outline of some computational challenges for EvoSysBio that motivated the development of a new programming language to mee these challenges.

 

Dr. Laurence Loewe
Genetics, UW-Madison
Fundamental Modes of Computing

While abstractions as scary as Turing machines, lambda calculus and solvers for conjunctive normal form statements have everything they need to turn away even the most patient biology student from giving computational approaches in biology a chance, it turns out that all programming languages combine some of the corresponding higher level concepts also known as imperative, functional and logic programming. Here a translation of these concepts into mor tangible language is presented that also extends includes concurrency and uses the physical world to re-interpret some of the basic questions that every fundamental mode of computing needs to answer about storage and processing.

 

Kate Scheuer, Jerdon Dresel, Laurence Loewe et al.
Genetics & WID, UW-Madison
Drosophila Circadian Clocks posters and FlyClockbase

Various posters along with a (large!) printout of flyclockbase will be presented to facilitate discussions of circadian clocks and related modeling questions.