Our past, present and future. Mostly present, as we are too young to remember and too old to think we know where we are heading.
Click a thumbnail for the relevant project to appear.
The evolution and genetic basis of cross-stress behavior
Project description
Living in complex environments that lead to exposures to fluctuations along many physical dimensions, bacteria have evolved diverse mechanisms to cope with such environmental variation. Severe fluctuations in such variables can be viewed as a form of stress and understanding responses to such stresses has been the focus of numerous studies in the past. Still, our knowledge of the underlying gene regulatory and biochemical networks that contribute to the diverse responses to stresses and their evolutionary potential to re-wire through adaptive evolution is limited. Even less is known about the coupling between responses and regulatory mechanisms when organisms are exposed to combinations of various stressors, as certainly is a common occurrence in natural environments. In certain cases, adaptation to one stressful environment provides a fitness (dis)advantage when cells are exposed to a second stressor, a phenomenon that has been coined as cross-stress protection. A tantalizing question in bacterial physiology is how this behavior emerges during adaptation and what the underlying mechanisms of acquired stress resistance are.
The goal of this project is to identify the genetic basis and evolutionary potential of cross-stress behavior. We use an integrative computational/experimental approach that has the following goals: (a) to provide insight on the genome-wide regulatory program that underlie stress responses, (b) to elucidate the underlying genes, pathways and regulatory sub-networks that are involved in microbial stress adaptation, (c) to comprehensively characterize cross-stress behavior in E. coli populations before and after evolution, (d) to identify the genetic basis of this phenomenon, (e) to identify stress pairs (A,B) where evolution in stress A renders the bacterial population more (less) protected when it gets exposed in stress B.
Lab members involved in the project
Related Publications
[J10] M. Dragosits, V. Mozhayskiy, S. Quinones-Soto, J. Park, I.Tagkopoulos, "Evolutionary potential,
cross-stress behavior, and the genetic basis of acquired stress resistance in Escherichia coli",
under review, Molecular Systems Biology, 2012
[J9] V. Mozhayskiy, I. Tagkopoulos, "Microbial evolution in vivo and in silico: methods and applications", under review, Integrative Biology, 2012
[J2] I.Tagkopoulos, Y.Liu, S. Tavazoie, "Predictive Behavior Within Microbial Genetic Networks",
Science, 320:1313-7, 2008
An integrated systems-level framework for automated circuit design
Project description
Recently, there have been notable advances in the field of computer-aided tools for synthetic circuit design at a systems level. While most approaches focus on providing visualization and simulation capabilities for the users to explore, there is a lack of an end-to-end integrated framework capable of automated optimization of synthetic circuits based on user-defined constraints. Such framework would be transformative in the design and construction of synthetic circuits that go beyond a handful of parts that is typical today.
A major direction to our laboratory is to develop an integrated framework towards the automated systems-level design of mixed-signal synthetic circuits. Towards this end, we employ and extend various techniques that include all-atom molecular dynamics (MD) simulations, mathematical optimization, graph-theoretic algorithms, and multiscale stochastic models of cellular processes. We use our experimental lab for part characterization and synthetic circuit validation. The main components of the framework that we have developed are summarized below:
- Part construction and characterization: We need well-characterized building blocks that will function as the fundamental units for building more complex circuits. We build mutant libraries of promoters, terminators and ribosome binding sites (RBS) with a large dynamic range in their activities. To predict the effect of mutation combinations and avoid the combinatorial explosion of an exhaustive search on all possible variations, we have constructed an all-atom method that minimizes statistical errors between simulation frames. So far we have used it to construct two mutant inducible promoter libraries that adhere to the biobricks standard and we have measured their activity in E. coli cells. Characterized components are stored in a database with other relational information (dependencies, compatibility with other parts, hosts, etc.).
- Optimization framework: We have developed and currently extend an optimization framework for the automated design of synthetic gene circuits. Given user-defined constraints, circuit topology and desired behavior, the optimization method provides the optimal set of components to be used. First, a module library that includes experimentally characterized constructs is queried through graph isomorphism methods to identify potential matches to the initial circuit topology. If matches have been identified, the graph is transformed to an equivalent sub-graph that then it is partitioned efficiently into sub-circuits. Then, an exact algorithm based on Mixed Integer Nonlinear Programming is applied to optimize the component set. The final (near) optimal circuits are simulated with a multi-scale simulator to account for stochastic effects.
- Circuit construction and validation: The optimized designs are then constructed and their dynamic behavior is experimentally tested in a variety of relevant environmental settings (media, temperature, host cell, chemostat cultivations, etc.). The final characterized circuits are then deposited in the module database to be re-used in the future. Parts of the proposed framework (mutant library and optimization) have already been used in the development of an experimental projects in our lab.
Lab members involved in the project
Related Publications
In preparation
An module-based, optimization solver for synthetic biology
Project description
C Computer-aided design of synthetic biological circuits is an active field where most efforts concentrate on creating an intuitive graphical user interface and a robust simulation engine that enables researchers to build biological systems and elucidate their dynamics. In terms of circuit optimization, the field is mostly relying in greedy algorithms such as simulated annealing and genetic algorithms. Although there is considerable merit in using these heuristics in exploring the vast solution space of synthetic gene design, these methods cannot guarrantee optimality, or bounds for the proposed solution(s). This project aims at bridging this gap by developing an optimization solver for synthetic biology using exact and approximation algorithms.
Lab members involved in the project
Related Publications
[J6] L. Huynh, J. Kececioglu, M. Koeppe, I.Tagkopoulos, "Automated Design of Synthetic Gene
Circuits through Linear Approximation and Mixed Integer Optimization", 7(4):e35529,
PLoS ONE, 2012
[C11] L. Huynh, I.Tagkopoulos, "A robust, library-based, optimization-driven method for automatic
gene circuit design", 2nd IEEE International Conference on Computational Advances
in Bio and Medical Sciences (ICCABS), Las Vegas, NV, pp.1-6, 24-26, 2012
[C11] L.Huynh, J.Kececioglu, I.Tagkopoulos, "Scaling responsibly, Towards a reusable, modular, automated gene circuit design", Proceedings of the 3rd International Workshop on Bio-design Automation, IWBDA'12, San Diego, 2012.
[C7] L.Huynh, J.Kececioglu, I.Tagkopoulos, "Automated Design of Synthetic Gene Circuits
through Linear Approximation and Mixed Integer Optimization", Proceedings of the 3rd
International Workshop on Bio-design Automation, IWBDA2011, San Diego, 2011.
A synthetic biology approach to self-regulatory recombinant protein production
Project description
Recombinant or heterologous protein production (RPP) is one of the most widely-used biotechnological process, with applications that range from catalysis (e.g. washing detergents) and therapeutic use (e.g. antibody-based therapeutics), to protein production for enzymatic characterization and crystallography. Unfortunately, high expression of the recombinant protein is detrimental to the cell and usually protein quality and quantity are inversely correlated at high product levels, as proteins misfold and inclusion bodies are formed.
To mitigate this problem, we developed a system where increased cellular stress, due to inadequate environmental conditions or over-production of the recombinant protein, leads to a decrease of RPP activity. Briefly, an inducible promoter is used to turn-on recombinant protein production. If and when the cell is stressed, a strong repressor that is under the control of a stress reporter is expressed, which shuts-down or decrease the production of the recombinant protein. Through this circuit, the host organism is capable of self-regulating the protein product as a stress-mediated response. In addition, we developed an ODE/DDE model to drive the design of the synthetic circuit and gain insight to the the system dynamics.
Lab members involved in the project
Martin Dragosits (postdoc alumni), Daniel Nicklas (rotating graduate student).
Related Publications
[J8] I. Tagkopoulos, "More control to the microbial factories: auto-regulatory control through engineered stress-induced feedback", accepted, Bioengineered, 2012
[J5] M. Dragosits, D. Nicklas, I.Tagkopoulos, "A synthetic biology approach to self-regulatory
recombinant protein production in Escherichia coli", 6:2, Journal of Biological Engineering,
2012
Engineering promoter libraries through computational prediction of gene expression
Project description
A necessary condition for any computational method to operate is the existence of underlying biological knowledge and parameter values. In the real of computer-aided design of biological systems, this translates to the existence of well-characterized parts with known function and a large dynamic range of operation. Towards this goal, we have created a computational and experimental pipeline to develop transcription factor-promoter libraries that can be used as biological blocks for the targeted engineering of biological circuits. We have developed a molecular dynamics model that can account for induced mutations in the promoter regions where known transcription factors bind, and we predict the change in the binding affinity through MD simulations. Then we build the promoter libraries through site-directed mutagenesis and characterize their dynamics through fluorescence measurements, shift assays and other tecniques.
Lab members involved in the project
Related Publications
Coming soon
