Quantitative computational models play an increasingly important role in modern biology. Such models typically involve many free parameters, and assigning their values is often a substantial obstacle to model development. Directly measuring in vivo biochemical parameters is difficult, and collectively fitting them to other experimental data often yields large parameter uncertainties. Nevertheless, in earlier work we showed in a growth-factor-signaling model that collective fitting could yield well-constrained predictions, even when it left individual parameters very poorly constrained. We also showed that the model had a ‘‘sloppy’’ spectrum of parameter sensitivities, with eigenvalues roughly evenly distributed over many decades. Here we use a collection of models from the literature to test whether such sloppy spectra are common in systems biology. Strikingly, we find that every model we examine has a sloppy spectrum of sensitivities. We also test several consequences of this sloppiness for building predictive models. In particular, sloppiness suggests that collective fits to even large amounts of ideal time-series data will often leave many parameters poorly constrained. Tests over our model collection are consistent with this suggestion. This difficulty with collective fits may seem to argue for direct parameter measurements, but sloppiness also implies that such measurements must be formidably precise and complete to usefully constrain many model predictions. We confirm this implication in our growth-factor-signaling model. Our results suggest that sloppy sensitivity spectra are universal in systems biology models. The prevalence of sloppiness highlights the power of collective fits and suggests that modelers should focus on predictions rather than on parameters.

https://www.lassp.cornell.edu/sethna/pubPDF/SloppyEverywhere.pdf

Universally sloppy parameter sensitivities in systems biology models.

https://digital.library.unt.edu/ark:/67531/metadc488175/m2/1/high_res_d/2003_phytochemistry_817.pdf

Plant metabolomics: large-scale phytochemistry in the functional genomics era

A fully compartmentalized genome-scale metabolic model of Saccharomyces cerevisiae that accounts for 750 genes and their associated transcripts, proteins, and reactions has been reconstructed and validated. All of the 1149 reactions included in this in silico model are both elementally and charge balanced and have been assigned to one of eight cellular locations (extracellular space, cytosol, mitochondrion, peroxisome, nucleus, endoplasmic reticulum, Golgi apparatus, or vacuole). When in silico predictions of 4154 growth phenotypes were compared to two published large-scale gene deletion studies, an 83% agreement was found between iND750's predictions and the experimental studies. Analysis of the failure modes showed that false predictions were primarily caused by iND750's limited inclusion of cellular processes outside of metabolism. This study systematically identified inconsistencies in our knowledge of yeast metabolism that require specific further experimental investigation.

/pdf/reconstruction-and-validation-of-saccharomyces-cerevisiae-1b6agx2944.pdf

Reconstruction and Validation of Saccharomyces cerevisiae iND750, a Fully Compartmentalized Genome-Scale Metabolic Model

Complex assemblies of interacting proteins carry out most of the interesting jobs in a cell, such as metabolism, DNA synthesis, movement and information processing. These physiological properties play out as a subtle molecular dance, choreographed by underlying regulatory networks. To understand this dance, a new breed of theoretical molecular biologists reproduces these networks in computers and in the mathematical language of dynamical systems.

/pdf/network-dynamics-and-cell-physiology-4ydb3fmmx2.pdf

Network dynamics and cell physiology

Eukaryotic cells have evolved various response mechanisms to counteract the deleterious consequences of oxidative stress Among these processes, metabolic alterations seem to play an important role We recently discovered that yeast cells with reduced activity of the key glycolytic enzyme triosephosphate isomerase exhibit an increased resistance to the thiol-oxidizing reagent diamide Here we show that this phenotype is conserved in Caenorhabditis elegans and that the underlying mechanism is based on a redirection of the metabolic flux from glycolysis to the pentose phosphate pathway, altering the redox equilibrium of the cytoplasmic NADP(H) pool Remarkably, another key glycolytic enzyme, glyceraldehyde-3-phosphate dehydrogenase (GAPDH), is known to be inactivated in response to various oxidant treatments, and we show that this provokes a similar redirection of the metabolic flux The naturally occurring inactivation of GAPDH functions as a metabolic switch for rerouting the carbohydrate flux to counteract oxidative stress As a consequence, altering the homoeostasis of cytoplasmic metabolites is a fundamental mechanism for balancing the redox state of eukaryotic cells under stress conditions

/pdf/dynamic-rerouting-of-the-carbohydrate-flux-is-key-to-ntdzqa95q1.pdf

Dynamic rerouting of the carbohydrate flux is key to counteracting oxidative stress

This paper examines whether the in vivo behavior of yeast glycolysis can be understood in terms of the in vitro kinetic properties of the constituent enzymes. In nongrowing, anaerobic, compressed Saccharomyces cerevisiae the values of the kinetic parameters of most glycolytic enzymes were determined. For the other enzymes appropriate literature values were collected. By inserting these values into a kinetic model for glycolysis, fluxes and metabolites were calculated. Under the same conditions fluxes and metabolite levels were measured. In our first model, branch reactions were ignored. This model failed to reach the stable steady state that was observed in the experimental flux measurements. Introduction of branches towards trehalose, glycogen, glycerol and succinate did allow such a steady state. The predictions of this branched model were compared with the empirical behavior. Half of the enzymes matched their predicted flux in vivo within a factor of 2. For the other enzymes it was calculated what deviation between in vivo and in vitro kinetic characteristics could explain the discrepancy between in vitro rate and in vivo flux.

Mike Schepper

Papers

Can yeast glycolysis be understood in terms of in vitro kinetics of the constituent enzymes? Testing biochemistry.

Glucose uptake kinetics and transcription of hxt genes in chemostat cultures of saccharomyces cerevisiae