We do research in the experimental, theoretical and the software/standards domains. The figure below summarizes our current efforts.

Wet Lab Research

Wilbert Copeland

I am developing in vivo tools to obtain quantitative measurements of gene expression dynamics in E. coli. Characterizing kinetic processes of gene networks is achieved by employing fluorescent RNA and protein molecules. Experimental measurements are linked to system models which describe the underlying biochemical reactions. Success within this research area will enable researchers to better probe the in vivo state of gene circuit components. Additionally, characterization data of synthetic gene networks will yield kinetic parameters that should increase the predictive capability of computational models.

Bryan Bartley

My research goal is to develop protocols for making precise, standardized measurements of the steady-state and dynamic performance of simple gene networks in E. coli.  The expression of fluorescent protein reporters in synthetic gene circuits is measured using multi-well spectrofluorimetry in E. coli populations, as well as single-cell methods such as fluorescence microscopy and flow cytometry.  One of the big challenges for making these standardized measurements is isolating the intrinsic properties of a gene circuit from its host context.  The performance of a synthetic gene circuit depends on the physiological state of the host cell, such as RNA polymerase, ribosome, and plasmid concentrations, all of which are regulated by E. coli according to conditions in the growth medium.  In addition, the expression of a non-native gene circuit in the host cell siphons away catabolic and energetic resources needed for growth, causing feedback effects to overall fitness and growth rate phenotype.  In order to disentangle the interaction between synthetic gene circuits and the host cell machinery, I am developing computational models of the stringent response in E. coli as well as using quantitative PCR to measure how physiology of the cell responds to different conditions in the growth medium.

Sean Sleight

My research involves studying the evolutionary stability of genetic circuits and metabolic pathways.  In particular, I have investigated the evolutionary stability dynamics and loss-of-function mutations in a variety of genetic circuits.  By understanding the mechanisms of loss-of-function mutations when using different combinations of genetic parts, I was able to re-engineer these circuits to be more mutationally robust over evolutionary time.  Instead of using a rational approach for robust circuit design, I am currently using a directed evolution approach to randomize large genetic circuits and metabolic pathways to select for increased mutational robustness.  To screen for robust designs, I am also developing a method that allows for a mutational “readout” device using color.  With these studies I hope to develop design principles for engineering robust synthetic systems.

 

Computational, Standards and Software

Michal Galdzicki

Research in systems and synthetic biology demands close collaboration between researchers and the ability to exchange information between software tools. Standards facilitate such research by enabling communication within workflows and re-use of information among different projects.  Exchange of information in the systems and synthetic biology fields relies on software standards, as simply relying on publications is not sufficient to completely reproduce or reuse results. Here in the Sauro Lab we believe new standards are needed to improve communication in the field and reduce the effort with which collaborations are established. We have been privileged to take leadership roles in the development of community based standards for both systems biology and synthetic biology. Over the years we have been involved in the development of

• SBML Systems Biology Markup
• SBGN Systems Biology Graphical Notation
• SEDML
• SBOL Synthetic Biology Open Language http://www.sbolstandard.org/

To build these standards we engage in collaborations with many stakeholders in the communities of interest and aim to build standards that are above all else useful. These computational standards will be able to capture the results of systems and synthetic biology experiments unambiguously. Software tools that support the standards will enable easy exchange of information. We hope this will improve the current capabilities of researchers and create new ones.

Jeff Johnson

Synthetic Biology Open Language: Today, biologists mainly exchange data in the form of ad-hoc diagrams or qualitative descriptions. But if we hope to engineer complex biological systems, we will need to develop an unambiguous standard for data exchange—one that can be read by computers as well as humans. The Synthetic Biology Open Language (SBOL) is a community effort to foster that transition. It provides SBOLv, a visual language analogous to circuit diagrams or architectural blueprints, for describing DNA constructs to people, and complementary serialization format for describing them to programs. We are developing libSBOLc, the C library that will read, write, and validate SBOL documents. Once it is completed, we will use it as the basis for a new desktop application that will make it easy to perform complex searches across multiple databases of available DNA constructs.

Emily Yang

My current project attempts to conduct a thorough in silico comparison between fluxes determined through flux balance analysis (FBA) and the steady-state fluxes determined via kinetic simulations.  I am developing a method for constructing realistic but synthetic networks.  Each network features randomly-generated reactions, including a biomass function around which the reaction rates are optimized.

 

Garrett Quesnell

Gillespie Stochastic Simulator in C: I am working on creating a Gillespie algorithm based stochastic simulator to model zero order and first order reactions. It uses a tau-leaping approach for simple computation and a Mersenne twister pseudorandom number generation for accurate results. The intent of this program is that it can be used as a light plugin to other programs to quickly estimate stochastic behavior.

 Hardware and Devices

Bennett Ng

Development of robust synthetic gene constructs requires insight into the evolutionary stability of such constructs. Evaluating evolutionary stability requires parallel cell culture experimentation which is time-consuming and laborious. Development of a low-cost, parallel chemostat device could help to accelerate this type of research. The proposed chemostat device features integrated heating, media flow, and imaging systems for fully-automated cell culture experiments. The device is constructed of common materials and components, and includes a reprogrammable Arduino microcontroller and an accompanying desktop computer application for customizable operation.

 

 

Herbert Sauro

Analog computer of the MAPK cascade with negative feedback. Controlled from a digital computer.

 

 

 

 

 

Bennett Ng and Bryan Bartley

Respressilator Circuit – coming soon

Rahul Brito

Lorenz Attractor Circuit – comming soon

Theoretical Work

Kyung Kim

 

My research specialty is in computational and mathematical analysis of stochastic and deterministic biological models, specializing in the field of systems and synthetic biology. Especially, in synthetic biology, there are important issues such as modularity, stochasticity, inter-cellular interactions and cellular communities, and evolutionary stability. I am investigating these issues by taking mathematical and computational approaches.

Modularity: Synthetic biology aims to engineer biological organisms with new functionalities by designing and constructing regulatory systems — synthetic circuits — from smaller network components. This bottom-up approach benefits from modularity. The input and output of the circuit components are commonly considered the concentration levels of regulatory proteins, e.g.,  transcription factors in gene regulatory networks and proteins in signal transduction networks. Here, I consider functional modularity, i.e., how well the functional relationship between the input-output (i/o) of circuit components is preserved, when the components are connected to another. I have introduced the well-known engineering concept of fan-out to gene regulatory networks to quantify the degree of modularity. The fan-out was shown to be closely related to retroactivity. I have found that the response time of transcription factors can linearly increase with the number of downstream promoter binding sites for the factors. The linearity was shown to hold for a wide class of transcription factor regulations. To measure the fan-out and retroactivity, gene expression noise was also shown to be useful. This theoretical prediction will be confirmed by experiments under the support of NSF in future.

Stochasticity: Intra-cellular processes such as transcription, translation, and cell cycles appear stochastic due to random biochemical reactions. These stochastic fluctuations, named gene expression noise, can significantly affect the phenotypes of host cells such as cellular differentiation and viral dynamics. For example, DNA uptake competence found in a soil bacterium, Bacillus subtilis, was known to be affected by the gene expression noise in ComK. In such noise-related phenotypes, it will be important to control the noise level with respect to its mean level, to distinguish the individual effect of the noise and mean levels and to understand the underlying biological mechanisms. Such noise control can be performed by identifying which system parameters to perturb and by how much. For this identification, I have developed stochastic control analysis (SCA) by extending metabolic control analysis to the stochastic regime and applied this analysis to yeast promoters, HIV-1 LTR promoters, and metabolic systems. In SCA, I found that there exist analogous summation theorems for the noise levels and they are related to time-scale separation in the network systems.

Other research projects related to inter-cellular interactions and cellular communities, mutational stability in synthetic circuits, and metabolic networks can be found in my personal website (link –http://faculty.washington.edu/kkim)

 

Software Development Best Practices

Code Repositories

Recommended repositories include sourceforge and Google code hosting. Use SVN where possible because it is the most widely used. The main page in a repository should include:

1. A description of the project
2. Direct link to the documentation (doxygen is the code implements an API)
3. Direct link to example code if the repository holds a library
4. How to obtain the source code (detailed)
5. Availability of Binaries
6. How to build the source on every supported platform (Use CMake where ever possible)
7. Instructions on how to update and build the documentation

The Simple Copasi API serves as a good example

Do’s and Don’ts for C/C++ Developers

1. Don’t use the Boost library, its too difficult to find experience programmers who can work effectively with Boost.

2. Only use the standard library when necessary. Use of Vector and string is however ok. 

3. Do not use C++ templates, or multiple inheritance and operator overloading

4. Don’t use Run-time type information (RTTI) as it is not yet well supported by all compilers.

5. Use Doxygen to document your code, particularly if it is a distributed library

6. Do not use TABS in your code, only spaces

7. Always comment your code