ilias tagkopoulos - US grants

This award is for a provisional allocation of time on the Blue Waters computer system, due to become operational in 2011, and for travel funds to support technical coordination by various collaborators with the Blue Waters project team and vendor technical team.

The project involves studies in bacterial ecology, gene regulatory networks and intracellular biochemical networks with a view to understanding the way in which populations of unicellular organisms evolve and adapt as their environment changes. This involves multi-scale biological systems where processes ranging from gene expression and intracellular biochemistry to ecosystem dynamics are in play. A key aim is to understand how, by considering the genetic and biochemical processes within a cell, unicellular organisms evolve to develop adaptive responses to recurring changes in their environments. The study will look at the influence of parameters such as nutrient concentrations and mutation on adaptation, compare the efficacy of different strategies for survival in static and fluctuating environments, and examine how unicellular organisms internalize the correlation structure of their environment by modifying their internal networks to facilitate such changes. The role of genetic and molecular information transfer processes will also be studied.

The modeling approach has been developed and previously used for scientific research using an implementation on contemporary computing systems. It is inherently multi-scale, including representations of molecular processes within cells and scaling up to ecosystems of unicellular organisms. In prior work, the limitations of computer power have necessitated a number of simplifying assumptions. It is anticipated that the use of Blue Waters will allow some of these to be relaxed, so that more biological processes can be included in the simulations. Results will be tested against the outcomes of in vivo experiments in the UC Davis Genome Center.

The principal investigator is a recent Ph.D. Progress in this research area could find application in bioengineering and biotechnology.

Synthetic Biology is a nascent field with applications that range from bio-fabrication to alternative energy. Despite its significance, engineering of biological circuits still relies on trial-and-error tinkering techniques, with limited computational support. If Synthetic Biology is to advance to more complex synthetic systems that go beyond a handful of interacting parts, a scalable, integrative, methodological approach is necessary. In an analogy to integrated circuits, when it comes to circuit engineering, the role of detailed computer models, optimization methods, simulators and design tools is paramount.

Intellectual Merit: This project aims to pave the way towards an optimization-based, automated design framework for synthetic gene circuits that adhere to user-defined constraints. A synthetic gene circuit is a collection of one or more genes, together with elements (promoters, ribosome binding sites, etc.) that influence gene expression. The wiring, i.e. the order and position of every element, within a synthetic gene circuit determines the gene expression pattern, and overall behavior of the circuit. These circuits are introduced, usually as part of a plasmid(s), in a host organism that can be readily manipulated in order to achieve a desired outcome (e.g. specific temporal behavior, or production of an enzyme).

To facilitate faster time-to-market solutions and more robust, predictable designs, PIs will develop a design and optimization tool prototype. To that end, PIs propose a new optimization formulation that encompasses multiple biological models relevant to synthetic genetic circuit design. In addition, they propose a hybrid optimization-simulation technique to capture additional effects related to cell division, noise, and evolutionary processes. The investigation will focus on how state-of-the-art techniques from combinatorial optimization can be applied to find the optimal circuit for a specific task. Since the tool will need a library of well-characterized components to operate, PIs will create a mutant library of three widely-used regulators, then quantitatively characterize them, and store this information in a publicly available database. As a proof-of-concept experiment, they will assess their integrative approach by constructing an automatically-designed synthetic circuit, measuring its output and deviation from the desired goal, and then comparing it to other similar designs that have been already available in literature.

Broader Impact: An optimization-based, design tool for synthetic biology has the potential to provide a service to the academic community by reducing drastically the time-to-market aspect of synthetic designs, and providing insight on biological function, thus accelerating research in an exponentially growing field. All components and characterized libraries that will be developed as part of this award will be publicly available, deposited in the synthetic biology community?s standard Parts Registry. Furthermore, this award will partially support the work and training of the UC Davis IGEM team, a synthetic biology undergraduate team who competes in the annual IGEM competition. Knowledge from this project will be directly transferred into classrooms through the course ECS 289K "Computational Challenges in Systems and Synthetic Biology" (UC Davis), and the course CSC 450/550 "Algorithms for Bioinformatics" (U. Arizona).

There have been recent and notable advances in the field of computer-aided tools for synthetic circuit design at the system level. While most approaches focus on providing visualization and simulation capabilities for the users to explore, there is a lack of a much needed end-to-end integrated framework capable of automated optimization of synthetic circuits based on user-defined constraints. The goal of the proposed project is to develop an integrated synthetic biology framework and tools for the automated systems-level design of mixed-signal synthetic circuits. This framework will serve as a computational tool for synthetic biologists and will lay the theoretical foundations for system level synthetic circuit design.

The project will explore a global optimization platform with both approximate and exact techniques for finding the optimal set of parts for any given configuration, which will provide guarantees on the optimality, reproducibility or bounds of the solution. This work will explore matching past validated designs to the circuit at hand, by employing graph-querying and graph partitioning algorithms that will transform the original design to equivalent graphs which can be solved efficiently. This will promote modular and reusable designs, and result in an increase in the level of achievable complexity and a decrease in the cost of construction.

Bacteria that are exposed to a stressful environment are often better protected when they are then exposed to a second stressor. Understanding of this behavior, which is also known as cross-stress protection, is limited especially when it comes to its genetic basis. This project will characterize the genetic basis and evolutionary potential of cross-stress behavior in the bacterium Escherichia coli. To elucidate the evolutionary potential of bacteria and the emergence of cross-stress protection, E. coli will be exposed to three stressors, and organisms will be identified in which evolution under a first stress has conferred a fitness advantage (or disadvantage) upon exposure to a second stress. DNA sequencing will be used to identify genetic changes whose phenotypic effects will be validated by reversing the mutations.

Broader Impacts: The project will generate new knowledge relevant to our understanding of microbial behavior in the agricultural and biotechnology sectors. The PI will host and train high school students in microbial evolution and bioinformatics through the UC Davis Biotechnology program. The PI will advise the UC Davis IGEM team in projects that include topics in microbial evolution and microbial stress response.

PI: S.P. Dinesh-Kumar (University of California-Davis)

CoPIs: Robert Gilbertson, Diane Ullman, and Ilias Tagkopoulos (University of California-Davis)

Tomato is the second most important vegetable crop in the world. Tomato production in the US and worldwide is threatened by two devastating insect-transmitted viruses - tomato spotted wilt virus (TSWV) and tomato yellow leaf curl virus (TYLCV). Management of these viruses is difficult and often relies heavily on the use of insecticides to control the vectors. Insecticidal control is often ineffective, has negative human health and environmental effects, and can lead to selection of insecticide-resistant forms of the vectors. In tomato, the Ty1 locus for TYLCV and the Sw5 locus for TSWV are major genetic resistance sources, and both have been introgressed into commercial tomato varieties. However, very little is known about the mechanisms involved in Ty1- and Sw5-mediated resistance. This project will provide new insights into the nature of and interplay between mechanisms of plant resistance and susceptibility to infection by insect-transmitted viruses. These studies will provide insights into whether similar or different pathways are targeted by RNA (TSWV) versus DNA (TYLCV) viruses during infection and/or host defense. The project will then use the results from these studies to optimize and validate a new approach called miRNA-induced gene silencing (MIGS) that can be used to assess gene function in tomato. If successful, the outcomes of this project will likely lead to a new generation of antiviral strategies against these viruses. The project will also provide a 10-week training program every year for undergraduate students from Sacramento City College and high school students from Davis Senior High School.

The genetic resistance mediated by Sw5 and Ty1 loci introgressed from wild tomato relatives into cultivated tomato provides the best source of protection against TSWV and TYLCV, respectively. The investigators of this project hypothesize that Sw5- and Ty1-mediated recognition of TSWV and TYLCV trigger early and discernable transcriptional changes that involve common and specific immune signaling networks. In addition, infection of these viruses on susceptible tomato genotypes induces transcriptional changes that promote virus virulence. To gain insights into the transcriptional changes, the investigators will use RNAseq approach to identify differentially expressed transcripts during TSWV and TYLCV resistance and susceptible responses in tomato. Computational methods will be used to analyze RNASeq data to identify key common and specific genes that are differentially regulated during TSWV and TYLCV resistance and susceptible interaction. The project will optimize and validate the MIGS approach for gene function studies in transgenic tomato by targeting known candidate genes. The optimized MIGS approach will be used to evaluate function of selected differentially expressed genes during TYLCV and TSWV resistance and susceptible interactions in tomato. Results from the project will likely provide insights into designing new generation of strategies to control not only TYLCV and TSWV but also other important plant pathogens. All data and biological resources will be accessible to the public. All RNAseq data will be deposited and available through GEO. All MIGS vector and candidate target gene constructs will be available upon request and through Addgene (https://www.addgene.org/) and ABRC (http://www.biosci.ohio-state.edu/~plantbio/Facilities/abrc/abrchome.htm). Seeds from transgenic tomato lines will be available through the Tomato Genetics Resource Center at UC Davis (http://tgrc.ucdavis.edu/).

This project aims to develop the next-generation of genome-scale, data-driven models for microbial organisms. The project will first focus on the most-studied microbe, the gram-negative bacterium Escherichia coli, due to the availability of high-throughput data, cellular organization, its significance to industry and human health. The project will take advantage of the multi-omics datasets that resulted from advances in parallel high-throughput molecular profiling over the past fifteen years, the emergence of data-driven, integrative, multi-scale models with substantial improvement of their predictive power and new techniques in machine learning, especially those related to deep learning. Accurate prediction of microbial fitness and cellular state can have profound implications to the way we test hypotheses that are directly related to health, social or economic benefits. This award will support the training of multiple undergraduate and graduate students in computational modeling and high-performance simulations of biological systems through undergraduate courses, IGEM teams and other initiatives.

This project will support the generation of knowledge from the largest normalized omics compendia for the most widely used microbe that will be a boon for the development and training of the next generation of data-driven predictive methods in molecular and cellular biology. It will provide the computational resources to evaluate a state-of-the-art multi-scale model with the capacity to predict phenotypic characteristics and environmental conditions from collective omics data. This will be the first systems-level simulator that targets a specific microbe (E. coli) and will be able to simulate populations of cells with a resolution ranging from individual gene concentrations to population dynamics. To achieve that, process migration, load-balancing and strong scaling techniques have to be adopted and applied in this context, which are all novel features for the area of whole-cell modeling. The proposed HPC simulations will be intimately related to hypothesis generation and testing. The simulations will address questions related to what their phenotype and expression profiles of microbial cultures are in complex environments. In the context of systems biology, integration of these techniques has the potential of being transformative, but only if the necessary computational infrastructure able to handle these tasks is available. The Blue Waters Supercomputer with its unique architecture, large-scale simulation capabilities and professional support staff provides the ideal platform to achieve this ambitious goal.

The goal of this project is to create novel methods that help researchers gather and integrate existing data sets in order to better inform the design of future experiments. Ideally new experiments include some replication, fill in gaps and produce new knowledge, but this balance is hard to achieve when existing data sets are in different locations or organized in very different ways. Thus to achieve the goal, two aims are proposed: 1) develop methods for creating cohesive biological datasets from public experiments such that they are suitable for training computational models; 2) develop methods that indicate what experimental conditions should be used to collect new datasets so that are the most likely to yield important information about the study organism's biological properties, like structure and behavior. Achieving this goal will enable an understanding of important rules of life for organisms more efficiently and economically, by focusing on the experiments that give us the most value for the funds spent.

This exploratory project will focus on data arising from genome-wide transcriptional profiling methods (e.g. microarrays, RNA-Seq), building a computational foundation for later expansion. First, optimal data processing techniques for creating integrated compendia will be assessed, in order to select the best method for building training datasets for machine learning methods. Second, data-driven computational models will be trained on the data compendia and evaluated for success in describing and microbial behavior. Third, given the normalized compendia (in the transcriptomics data space) an optimal experimental design methodology will be prototyped, to recommend the best set of experiments to perform to yield the complete set of data needed to fit and test the biological model. The experimental design methodology will be benchmarked using synthetic data, and then evaluated by exploring the effect of design- recommended combinations of antibiotics and antiseptics (10 in all) on microbial behavior. This will be compared to the outcomes of experiments designed by methods currently used. Success metrics will focus on how quickly the required information in the experimental space is gathered and what level of uncertainty in a model remains after each experiment is completed.

PROJECT SUMMARY Depression is an impairing and prevalent mood disorder in adolescence, affecting 1 in 6 youth by age 18. So far, the use of conventional statistical approaches has had limited success in delivering tools for accurate individualized prediction of future depression for a specific child. The objective of this project is to build advanced non-linear machine learning algorithms integrating information from multiple sources to deliver accurate, individualized prediction. To accomplish this objective, the research team will use 5,000 variables (biological, cognitive, socio-emotional, environmental) measured on multiple occasions between the prenatal period and age 10 in children from the Avon Longitudinal Study of Parents and Children (ALSPAC, N = 15,636). The goal is to use features from the prenatal period to age 10 to estimate risk of reaching clinical levels of depressive symptoms between the ages of 12 and 18 years old. The research team will pursue two specific aims. The first aim is to build an algorithm for accurate prediction of adolescent depression by using informative features from the prenatal period until age 10 with machine learning methods that capture complex, multi-variate associations. The team will use several techniques, including artificial neural networks that exploit temporal information (recurrent neural networks, Long Short-Term Memory networks) to identify constellations of highly predictive features. Based on early-life stress sensitization theory, the first hypothesis is that features from the prenatal and early postnatal periods (up to age 5) provide greater predictive power than features from ages 6-10. The second aim is to determine if features predicting depression are unique to depression or shared with anxiety disorder and substance use disorder. Machine learning algorithms will predict age 18 clinical diagnoses of depression, anxiety disorder, and substance use disorder. The team will test the second hypothesis that some predictive features will be unique for each disorder and some will be shared across all three disorder types (e.g., childhood trauma). By accomplishing these aims, the research team will devise a clinically useful algorithm to estimate a child?s probability of developing adolescent depression. All software that will be created for this project will be open-source, and made freely available online in public repositories. Algorithms that would allow accurate early identification of children at risk to develop depression during future adolescent years would provide new avenues for preemptive interventions. This would yield enormous public health benefits by prioritizing treatment and shifting developmental trajectories away from eventual disorder for millions of individuals worldwide. To realize the potential of this overall impact on the field and society, predictive models that calculate risk with high sensitivity and specificity in childhood are needed. The proposed project aims to use robust, rigorous machine learning algorithms to take on this challenge.