Integrating quantitative proteomics and metabolomics with a genome-scale metabolic network model - PubMed (original) (raw)

Integrating quantitative proteomics and metabolomics with a genome-scale metabolic network model

Keren Yizhak et al. Bioinformatics. 2010.

Abstract

Motivation: The availability of modern sequencing techniques has led to a rapid increase in the amount of reconstructed metabolic networks. Using these models as a platform for the analysis of high throughput transcriptomic, proteomic and metabolomic data can provide valuable insight into conditional changes in the metabolic activity of an organism. While transcriptomics and proteomics provide important insights into the hierarchical regulation of metabolic flux, metabolomics shed light on the actual enzyme activity through metabolic regulation and mass action effects. Here we introduce a new method, termed integrative omics-metabolic analysis (IOMA) that quantitatively integrates proteomic and metabolomic data with genome-scale metabolic models, to more accurately predict metabolic flux distributions. The method is formulated as a quadratic programming (QP) problem that seeks a steady-state flux distribution in which flux through reactions with measured proteomic and metabolomic data, is as consistent as possible with kinetically derived flux estimations.

Results: IOMA is shown to successfully predict the metabolic state of human erythrocytes (compared to kinetic model simulations), showing a significant advantage over the commonly used methods flux balance analysis and minimization of metabolic adjustment. Thereafter, IOMA is shown to correctly predict metabolic fluxes in Escherichia coli under different gene knockouts for which both metabolomic and proteomic data is available, achieving higher prediction accuracy over the extant methods. Considering the lack of high-throughput flux measurements, while high-throughput metabolomic and proteomic data are becoming readily available, we expect IOMA to significantly contribute to future research of cellular metabolism.

PubMed Disclaimer

Figures

Fig. 1.

Fig. 1.

The figure illustrates the associations between variables in IOMA's optimization problem imposed by the various constraints. Rows represent enzymes and columns represent growth conditions—i.e. the _j_-th column representing the flux distribution for the _j_-th condition (denoted by formula image. Mass-balance [Equation (6)] and reaction directionality [Equation (7)] create dependencies between fluxes through different enzymes in one condition, irrespectively of all others conditions (i.e. associating fluxes in one column). The enzyme-kinetic constraint [Equation (8)] associates between fluxes through one enzyme in different growth conditions (via the enzyme's parameters _v_max+ and _v_max−, which are condition-invariant), irrespectively of all other enzymes (i.e. associating fluxes in one row). The latter constraint is defined only for a core set of enzymes for which metabolomic and proteomic data is available (marked in blue). The union of both types of row and column constraints in IOMA's optimization indirectly associates between many additional fluxes through various enzymes in different conditions.

Fig. 2.

Fig. 2.

Precision, recall and accuracy of predicted changes in fluxes between the wild-type and knockout strains in the RBC model, obtained by IOMA (blue) and MOMA (red). The average and standard deviation of the precision, recall and accuracy are shown across the 50 simulation runs in two scenarios. (a) No flux data is given as input for MOMA and IOMA, hence MOMA relies on random sampling of possible wild-type flux distributions to predict knockout effects. (b) Exchange fluxes are given as input to both methods, and are used by MOMA to obtain a more reliable prediction of the wild-type flux distribution. In both test scenarios, IOMA's predictions are significantly more accurate.

Fig. 3.

Fig. 3.

Precision, recall and accuracy of predicted changes in flux between wild-type and following gene knockouts in the E.coli model, obtained by IOMA (blue), FBA (green) and MOMA (red).

Fig. 4.

Fig. 4.

An example flux distribution predicted by IOMA following the knockout of talB (Transaldolase B). The network shows E.coli's glycolysis and PPP. Green (red) edges represent a measured increase (decrease) of flux between the wild-type and knockout strains. The letters F, M and I, represent predictions made by FBA, MOMA and IOMA, respectively, with green (red), representing a predicted increase (decrease) in flux. Predictions of no significant change in flux are not shown. As evident, IOMA correctly predicts the measured pattern of increased flux throughout glycolysis and decreased flux throughout the PPP pathways, with only two mismatches, while FBA and MOMA perform significantly worse.

References

    1. Åkesson M, et al. Integration of gene expression data into genome-scale metabolic models. Metabolic Eng. 2004;6:285–293. -PubMed
    1. Becker SA, Palsson BO. Context-specific metabolic networks are consistent with experiments. PLoS Comput. Biol. 2008;4:e1000082. -PMC -PubMed
    1. Becker SA, et al. Quantitative prediction of cellular metabolism with constraint-based models: the COBRA toolbox. Nature Protocols. 2007;2:727–738. -PubMed
    1. Bennett BD, et al. Absolute metabolite concentrations and implied enzyme active site occupancy in Escherichia coli. Nat. Chem. Biol. 2009;5:593–599. -PMC -PubMed
    1. Blank LM, et al. Metabolic-flux and network analysis in fourteen hemiascomycetous yeasts. FEMS Yeast Res. 2005;5:545–558. -PubMed

Publication types

MeSH terms

LinkOut - more resources