skip to content

Timetable (UMCW05)

Stoichiometric modelling (SM) of microbial metabolism

Tuesday 4th November 2014

Tuesday 4th November 2014
10:00 to 10:05 Welcome from Orkun Soyer INI 1
10:05 to 10:20 Taking a bottom-up approach: using metabolic models to study how interactions shape natural-occurring microbial communities
Co-authors: Rafi Zareki (Tel Aviv University), Omer Eilam (Tel Aviv University), Eytan Ruppin (Tel Aviv University)

Revealing the ecological principles that shape communities is a major challenge of the post-genomic era. To date, a systematic approach for describing inter-species interactions has been lacking. Using CBM, we independently predict the competitive and cooperative potential between 6,903 bacterial pairs derived from a collection of 118 species ’ metabolic models and chart an intricate association between competition and cooperation. Utilizing ecological data from 2,801 samples, we explore the associations between bacterial interactions and coexistence patterns. The high level of competition observed between species with mutual-exclusive distribution patterns supports the role of competition in community assembly. Cooperative interactions are typically unidirectional with no obvious benefit to the giver. However, within their natural communities, bacteria typically form close cooperative loops resulting in indirect benefit to all species involved.

10:20 to 10:35 tba INI 1
10:35 to 10:50 D Fell (Oxford Brookes University)
Issues in Flux Balance Analysis
Co-authors: Mark Poolman (Oxford Brookes University), Hassan Hartman (Oxford Brookes University)

As the practice of genome scale metabolic modelling by FBA has become more widespread, certain procedures and assumptions have been automatically adopted, almost as a standard, whereas their utility and applicability should be assessed for each specific model and investigation. In addition, there are some recurrent biochemical errors that are not always being filtered out.

Amongst the procedures not being given sufficient thought are: 1. Optimisation by maximisation or minimisation? Maximisation is generally adopted even though the way that the linear programming algorithm operates results in artefacts in the solutions that are not present on minimisation and that require post-processing to remove. 2. Expressing the biomass formation as a pseudo-reaction with non-integer stoichiometry or as part of the constraints in the analysis. Again the former is more general even though it is almost impossible to ensure it is correctly stoichiometrically balanced and feasible. 3. Substrate consumption for non-growth associated cell maintenance is for ATP generation. We have shown that part of this extra substrate consumption in Arabidopsis cells is for NADPH, presumably to combat oxidative damage, and this has consequences for the predicted fluxes in central carbon metabolism.

Recurrent errors include: 1. Writing enzyme prosthetic groups as substrates and products of enzyme reactions. FAD and FADH are the most frequent culprits. This creates pool metabolites that could generate spurious redox interactions across the network that will not exist because these groups are contained and recycled entirely within the enzyme reaction. 2. ATP from nothing. There are published models that can generate the ATP to satisfy maintenance requirements without any flows into the metabolic network from external material. Needless to say, all subsequent analysis of such a model is valueless. Preventing this is an elementary reality check during model construction.

10:50 to 11:15 Morning Coffee and Discussions
11:15 to 11:30 Importance of Standardising Genome-Scale Stoichiometric Models
Co-author: Orkun Soyer (University of Warwick)

Genome-scale stoichiometric model based approaches are increasingly being used for studying cellular metabolism, which has resulted in the development of stoichiometric models for a verity of organisms and cell types. However, many of these models vary significantly in terms of their syntax, chemical/reaction naming conventions, data structures, etc., mostly due to the use of different tools/methods for developing them. These variations often make the models incompatible; therefore, comparison/integration of multiple models become very difficult. In this talk I discuss about certain issues related to the lack of a common standard and emphasis the importance of standardising stoichiometric models.

11:30 to 11:45 Standardisation of stoichiometric models: how and why
Interest in constraint-based modelling of metabolism using stoichiometric models has grown significantly over the last 10-15 years. Hundreds of curated models [1], and thousands of automatically generated models [2] are now publicly available, covering organisms in all three domains.

Despite attempts of standardising their representation, using community-developed formats such as the Systems Biology Markup Language, SBML [3], many tasks surrounding model building and analysis are hampered by a lack of interoperability between models.

Based on the speaker's experience in co-leading two large international community efforts in the development of consensus models for yeast [4] and human [5], approaches to model standardisation will be discussed. Moreover, the benefits of adopting a disciplined approach to model standardisation - automated model building, model checking, and 'omics data integration - will be demonstrated.

Such reliance on automated techniques will be of particular relevance to stoichiometric modelling of microbial communities, where the complexity of such models is likely to far exceed that of even the largest existing models of mammalian metabolism.

[1] Optimizing genome-scale network reconstructions. Monk J, et al. Nat Biotechnol. 2014, 32(5):447-52. [2] Path2Models: large-scale generation of computational models from biochemical pathway maps. Büchel F, et al. BMC Syst Biol. 7:116. [3] The systems biology markup language (SBML): a medium for representation and exchange of biochemical network models. Hucka M, et al. Bioinformatics. 2003, 19(4):524-31. [4] A consensus yeast metabolic network reconstruction obtained from a community approach to systems biology. Herrgård MJ, et al. Nat Biotechnol. 2008, 26(10):1155-60. [5] A community-driven global reconstruction of human metabolism. Thiele I, et al. Nat Biotechnol. 2013, 31(5):419-25.

11:45 to 12:00 A Heinken (Université du Luxembourg)
Constraint-based modeling of microbial communities and their interaction with the host
Co-author: Ines Thiele (Luxembourg Centre for Systems Biomedicine)

The gut microbiota performs a central role in human well-being and disease, yet few constraint-based efforts have investigated the metabolic interaction between the host and this important ecosystem. In a first effort, we constructed an in silico model of a gnotobiotic mouse colonized with a single microbial species by joining a genome-scale, manually curated and validated reconstruction of the prominent human and mouse gut symbiont Bacteroides thetaiotaomicron and a previously published mouse reconstruction, iMM1415. We depicted the trade-off between host and microbe biomass and predicted that the microbe rescued lethal host phenotypes (1). Subsequently, we developed a framework of alternating in silico and in vitro steps to elucidate the metabolic potential of a poorly studied gut microbe relevant for human health, Faecalibacterium prausnitzii. The combination of metabolic modeling and in vitro culture revealed novel carbon sources and secretion products and allowed the definition of a chemically defined growth medium for the microbe (2). We then constructed an in silico gut microbial community model consisting of 11 microbes spanning three phyla. The community model was joined with the global human reconstruction Recon2 and the effects of the microbes on human metabolic phenotypes were systematically predicted. The microbes had a global effect on the predicted host body fluid metabolome that was significantly more pronounced for commensal than for pathogenic microbes. Finally, we discuss challenges that need to be considered when constructing a well-curated, representative gut microbial community model. We demonstrate that constraint-based multi-species modeling can accurately capture the interaction between the gut microbiota and the human host and should prove useful for modeling the influence of the microbiota in human health and disease.

1. A. Heinken et al., Gut Microbes 4, 28-40 (2013). 2. A. Heinken et al., J Bacteriol, (2014).

12:00 to 13:00 Sandwich Lunch at INI
13:00 to 13:15 B Olivier (Vrije Universiteit Amsterdam)
Exchanging stoichiometric models: interoperability at genome scale
Advances in the methods used to construct genome scale constraint based models and the wider adoption of constraint based modeling in biotechnological and medical applications have led to a rapid increase in both the number of models being constructed and the tools used to analyze them.

Faced with such growth, both in number and diversity, the need for a standardized data format for the definition, exchange and annotation of constraint based models has become critical. As the core model components (e.g. species, reactions, stoichiometry) can already be efficiently described in the Systems Biology Markup Language (SBML) the Flux Balance Constraints (FBC) package aims to extend SBML Level 3 core by adding the elements necessary to encode current and future constraint based model.

In this presentation, focusing on how they might be utilized in modelling microbial ecosystems, I will introduce the SBML Level 3 FBC package and two related standards, SED-ML and the COMBINE archive.

I will also explain the need for the consistent use of reaction/species identifiers and a tool that is being developed to address this issue.

13:15 to 13:30 tba INI 1
13:30 to 13:45 Optimal metabolic dynamics resolved by cyclic FBA
Co-authors: Willi Gottstein (Humboldt University Berlin, Institute of Theoretical Biology, Germany), Rainer Machne (Charité Universitätsmedizin Berlin, Institute of Theoretical Biology, Berlin, Germany), Ralf Steuer (Humboldt University Berlin, Institute of Theoretical Biology, Germany), Douglas Murray (Keio University, Institute for Advanced Biosciences, Japan)

We present a new variant of Flux Balance Analysis (FBA) that allows for dynamic modelling of metabolic cycles. In contrast to traditional dynamic FBA, which performs a series of simulations, discrete time frames are optimized simultaneously. The network is expanded to represent all species in every time frame and time transition fluxes are added to allow an exchange of metabolites between adjacent time frames. The steady state condition only holds true for the total process, and storage or depletion of metabolites throughout the cycle can be predicted. We apply this method to respiration cycles of continous yeast cultures to evaluate preferences of metabolic syntheses towards different energetic cellular states.

13:45 to 14:00 Implementing Trade-offs in FBA
Co-author: Orkun Soyer (University of Warwick)

The implementation of higher level constraints in flux balance analysis, that impose global limits on flux rates, could considerably limit the amount of apriori kinetic information that needs to go into an FBA model. Two promising approaches in this direction are the notion of "molecular crowding", that sets a limit on the total allowable flux within a model and the membrane-space limitation approach, that limits total uptake flux. We use membrane-space limitation in an evolutionary scenario to find uptake fluxes on FBA models that lead to higher growth rates given a certain medium. Depending on the complexity of the medium, the trade offs forbid a unique solution and hence we see the emergence of multiple co-existing model "species" sharing a given medium.

14:00 to 14:30 Afternoon Tea and Discussions
14:30 to 17:00 Discussions (suggested topics): Limitations of SM, Optimisation function, Standardisation of SM models INI 1
University of Cambridge Research Councils UK
    Clay Mathematics Institute London Mathematical Society NM Rothschild and Sons