Erroneous energy-generating cycles in published genome scale metabolic networks: Identification and removal

Abstract

Energy metabolism is central to cellular biology. Thus, genome-scale models of heterotrophic unicellular species must account appropriately for the utilization of external nutrients to synthesize energy metabolites such as ATP. However, metabolic models designed for flux-balance analysis (FBA) may contain thermodynamically impossible energy-generating cycles: without nutrient consumption, these models are still capable of charging energy metabolites (such as ADP→ATP or NADP+→NADPH). Here, we show that energy-generating cycles occur in over 85% of metabolic models without extensive manual curation, such as those contained in the ModelSEED and MetaNetX databases; in contrast, such cycles are rare in the manually curated models of the BiGG database. Energy generating cycles may represent model errors, e.g., erroneous assumptions on reaction reversibilities. Alternatively, part of the cycle may be thermodynamically feasible in one environment, while the remainder is thermodynamically feasible in another environment; as standard FBA does not account for thermodynamics, combining these into an FBA model allows erroneous energy generation. The presence of energy-generating cycles typically inflates maximal biomass production rates by 25%, and may lead to biases in evolutionary simulations. We present efficient computational methods (i) to identify energy generating cycles, using FBA, and (ii) to identify minimal sets of model changes that eliminate them, using a variant of the GlobalFit algorithm.

On the deformability of an empirical fitness landscape by microbial evolution

“…We used this dataset to create a "universal" metabolic model that included all reactions found in E. coli iJO1366 as well as a set of all potential novel reactions. We removed reactions that would lead to erroneous energy-generating cycles using the ModelFit algorithm (Fritzemeier et al, 2017). The algorithm was constrained to conserve reactions present in the original E. coli model.…”

Engineering the metabolic pathways of lipid biosynthesis to develop robust microalgal strains for biodiesel production

“…ModelSEED and PATRIC work on RAST (Rapid Annotations using Subsystems Technology) that contrast the genomic information against the already available information of phylogenetic neighbors . However, rapid reconstructions by using these automated tools are observed to have high mistake rate, so this is truly important to go for quality control and quality assurance (QC/QA) test, especially to metabolite being produced or utilized within the network in its particular reaction and energy production with no input . Models reconstructed through automated or semi‐automated means thus need intensive manual curation to ensure the high quality, so correct prediction can be made.…”

Memote: A community driven effort towards a standardized genome-scale metabolic model test suite

“…This behavior occurs when a model's reaction directions are not checked for thermodynamic feasibility, leading to the formation of flux cycles which provide reduced metabolites to the model without requiring nutrient uptake. Fritzemeier et al 10 detected such erroneous energy-generating cycles (EGCs) in the majority of GEMs specifically in the MetaNetX 11,12 (~66%) and ModelSEED 13 (~95%) databases, which mostly contain automatically-generated, non-curated metabolic models. Although the authors found that EGCs are rare in manually-curated GEMs from the BiGG Models database (~4%), their effect on the predicted growth rate in FBA may account for an increase of up to 25%.…”

“…Toolbox14 , or the modified GlobalFit algorithm15 presented by Fritzemeier et al10 . Yet, as the models of P. putida analyzed byYuan et al show, this is not done consistently 9 .…”