Limits...
Comparative Analysis of Yeast Metabolic Network Models Highlights Progress, Opportunities for Metabolic Reconstruction.

Heavner BD, Price ND - PLoS Comput. Biol. (2015)

Bottom Line: We have also compared pairwise gene knockout essentiality predictions for 10 of these models.We found that varying approaches to model scope and annotation reflected the involvement of multiple research groups in model development; that single-gene essentiality predictions were affected by simulated medium, objective function, and the reference list of essential genes; and that predictive ability for single-gene essentiality did not correlate well with predictive ability for our reference list of synthetic lethal gene interactions (R = 0.159).We conclude that the reconstruction of the yeast metabolic network is indeed gradually improving through the iterative process of model development, and there remains great opportunity for advancing our understanding of biology through continued efforts to reconstruct the full biochemical reaction network that constitutes yeast metabolism.

View Article: PubMed Central - PubMed

Affiliation: Institute for Systems Biology, Seattle, Washington, United States of America.

ABSTRACT
We have compared 12 genome-scale models of the Saccharomyces cerevisiae metabolic network published since 2003 to evaluate progress in reconstruction of the yeast metabolic network. We compared the genomic coverage, overlap of annotated metabolites, predictive ability for single gene essentiality with a selection of model parameters, and biomass production predictions in simulated nutrient-limited conditions. We have also compared pairwise gene knockout essentiality predictions for 10 of these models. We found that varying approaches to model scope and annotation reflected the involvement of multiple research groups in model development; that single-gene essentiality predictions were affected by simulated medium, objective function, and the reference list of essential genes; and that predictive ability for single-gene essentiality did not correlate well with predictive ability for our reference list of synthetic lethal gene interactions (R = 0.159). We conclude that the reconstruction of the yeast metabolic network is indeed gradually improving through the iterative process of model development, and there remains great opportunity for advancing our understanding of biology through continued efforts to reconstruct the full biochemical reaction network that constitutes yeast metabolism. Additionally, we suggest that there is opportunity for refining the process of deriving a metabolic model from a metabolic network reconstruction to facilitate mechanistic investigation and discovery. This comparative study lays the groundwork for developing improved tools and formalized methods to quantitatively assess metabolic network reconstructions independently of any particular model application, which will facilitate ongoing efforts to advance our understanding of the relationship between genotype and cellular phenotype.

Show MeSH

Related in: MedlinePlus

Growth simulations demonstrate interplay between network reconstruction and constraints.A) Optimal biomass flux calculated by flux balance analysis increased linearly with glucose uptake flux for all models when the glucose exchange reaction is the only constrained media component. All model predictions had a 0.8158 correlation with previously reported measured growth rate. B) When glucose and oxygen exchange reactions were both constrained to experimental values, there are high-correlation (black) and low-correlation models (red). C) Restricting flux through a mitochondrial aspartate transport reaction did not affect the predictions for the high correlation models, and improved all remaining correlations to >0.9, with the exception of the Yeast 4 model, which still over-predicted the maximum biomass flux at high glucose:oxygen exchange constraint ratios.
© Copyright Policy
Related In: Results  -  Collection

License
getmorefigures.php?uid=PMC4643975&req=5

pcbi.1004530.g006: Growth simulations demonstrate interplay between network reconstruction and constraints.A) Optimal biomass flux calculated by flux balance analysis increased linearly with glucose uptake flux for all models when the glucose exchange reaction is the only constrained media component. All model predictions had a 0.8158 correlation with previously reported measured growth rate. B) When glucose and oxygen exchange reactions were both constrained to experimental values, there are high-correlation (black) and low-correlation models (red). C) Restricting flux through a mitochondrial aspartate transport reaction did not affect the predictions for the high correlation models, and improved all remaining correlations to >0.9, with the exception of the Yeast 4 model, which still over-predicted the maximum biomass flux at high glucose:oxygen exchange constraint ratios.

Mentions: The “C-limited” simulations reflected a different behavior. When we constrained the glucose exchange reaction alone, all models had a 0.816 correlation with the reported growth rates (Fig 6A). However, the growth rates labeled “C-limited growth aerobic” by Österlund et al. are not linear over the range of constraints imposed on the glucose exchange reaction, suggesting that carbon (glucose) flux is not the sole growth-limiting factor, particularly at the higher range of glucose flux constraints. The ratio of glucose exchange flux to oxygen exchange flux would be expected to strongly influence maximum achievable biomass flux due to stoichiometric constraints on the oxidation of glucose [49]. We tested model behavior against this expectation by conducting FBA with both glucose and oxygen exchange reactions constrained to values reported by Österlund et al. [37]. When glucose and oxygen exchange reactions were both constrained to experimental values, we observed that the models segregated to 2 groups: biomass flux predictions made by 7 models (iFF708, iIN800, Yeast 5, iTO977, iMM904, and iMM904bs) correlated with observations with a correlation >0.9, and predictions made by the remaining models (Yeast 4, Yeast 6, Yeast 7, iAZ900) had lower correlations (Fig 6B).


Comparative Analysis of Yeast Metabolic Network Models Highlights Progress, Opportunities for Metabolic Reconstruction.

Heavner BD, Price ND - PLoS Comput. Biol. (2015)

Growth simulations demonstrate interplay between network reconstruction and constraints.A) Optimal biomass flux calculated by flux balance analysis increased linearly with glucose uptake flux for all models when the glucose exchange reaction is the only constrained media component. All model predictions had a 0.8158 correlation with previously reported measured growth rate. B) When glucose and oxygen exchange reactions were both constrained to experimental values, there are high-correlation (black) and low-correlation models (red). C) Restricting flux through a mitochondrial aspartate transport reaction did not affect the predictions for the high correlation models, and improved all remaining correlations to >0.9, with the exception of the Yeast 4 model, which still over-predicted the maximum biomass flux at high glucose:oxygen exchange constraint ratios.
© Copyright Policy
Related In: Results  -  Collection

License
Show All Figures
getmorefigures.php?uid=PMC4643975&req=5

pcbi.1004530.g006: Growth simulations demonstrate interplay between network reconstruction and constraints.A) Optimal biomass flux calculated by flux balance analysis increased linearly with glucose uptake flux for all models when the glucose exchange reaction is the only constrained media component. All model predictions had a 0.8158 correlation with previously reported measured growth rate. B) When glucose and oxygen exchange reactions were both constrained to experimental values, there are high-correlation (black) and low-correlation models (red). C) Restricting flux through a mitochondrial aspartate transport reaction did not affect the predictions for the high correlation models, and improved all remaining correlations to >0.9, with the exception of the Yeast 4 model, which still over-predicted the maximum biomass flux at high glucose:oxygen exchange constraint ratios.
Mentions: The “C-limited” simulations reflected a different behavior. When we constrained the glucose exchange reaction alone, all models had a 0.816 correlation with the reported growth rates (Fig 6A). However, the growth rates labeled “C-limited growth aerobic” by Österlund et al. are not linear over the range of constraints imposed on the glucose exchange reaction, suggesting that carbon (glucose) flux is not the sole growth-limiting factor, particularly at the higher range of glucose flux constraints. The ratio of glucose exchange flux to oxygen exchange flux would be expected to strongly influence maximum achievable biomass flux due to stoichiometric constraints on the oxidation of glucose [49]. We tested model behavior against this expectation by conducting FBA with both glucose and oxygen exchange reactions constrained to values reported by Österlund et al. [37]. When glucose and oxygen exchange reactions were both constrained to experimental values, we observed that the models segregated to 2 groups: biomass flux predictions made by 7 models (iFF708, iIN800, Yeast 5, iTO977, iMM904, and iMM904bs) correlated with observations with a correlation >0.9, and predictions made by the remaining models (Yeast 4, Yeast 6, Yeast 7, iAZ900) had lower correlations (Fig 6B).

Bottom Line: We have also compared pairwise gene knockout essentiality predictions for 10 of these models.We found that varying approaches to model scope and annotation reflected the involvement of multiple research groups in model development; that single-gene essentiality predictions were affected by simulated medium, objective function, and the reference list of essential genes; and that predictive ability for single-gene essentiality did not correlate well with predictive ability for our reference list of synthetic lethal gene interactions (R = 0.159).We conclude that the reconstruction of the yeast metabolic network is indeed gradually improving through the iterative process of model development, and there remains great opportunity for advancing our understanding of biology through continued efforts to reconstruct the full biochemical reaction network that constitutes yeast metabolism.

View Article: PubMed Central - PubMed

Affiliation: Institute for Systems Biology, Seattle, Washington, United States of America.

ABSTRACT
We have compared 12 genome-scale models of the Saccharomyces cerevisiae metabolic network published since 2003 to evaluate progress in reconstruction of the yeast metabolic network. We compared the genomic coverage, overlap of annotated metabolites, predictive ability for single gene essentiality with a selection of model parameters, and biomass production predictions in simulated nutrient-limited conditions. We have also compared pairwise gene knockout essentiality predictions for 10 of these models. We found that varying approaches to model scope and annotation reflected the involvement of multiple research groups in model development; that single-gene essentiality predictions were affected by simulated medium, objective function, and the reference list of essential genes; and that predictive ability for single-gene essentiality did not correlate well with predictive ability for our reference list of synthetic lethal gene interactions (R = 0.159). We conclude that the reconstruction of the yeast metabolic network is indeed gradually improving through the iterative process of model development, and there remains great opportunity for advancing our understanding of biology through continued efforts to reconstruct the full biochemical reaction network that constitutes yeast metabolism. Additionally, we suggest that there is opportunity for refining the process of deriving a metabolic model from a metabolic network reconstruction to facilitate mechanistic investigation and discovery. This comparative study lays the groundwork for developing improved tools and formalized methods to quantitatively assess metabolic network reconstructions independently of any particular model application, which will facilitate ongoing efforts to advance our understanding of the relationship between genotype and cellular phenotype.

Show MeSH
Related in: MedlinePlus