CoChat

The DOI matches its Crossref record; 66 of the 66 references it cites were found in a registry or an index.

Bayesian genome scale modelling identifies thermal determinants of yeast metabolism

Open access

Published version of a preprint

Abstract

The molecular basis of how temperature affects cell metabolism has been a long-standing question in biology, where the main obstacles are the lack of high-quality data and methods to associate temperature effects on the function of individual proteins as well as to combine them at a systems level. Here we develop and apply a Bayesian modeling approach to resolve the temperature effects in genome scale metabolic models (GEM). The approach minimizes uncertainties in enzymatic thermal parameters and greatly improves the predictive strength of the GEMs. The resulting temperature constrained yeast GEM uncovers enzymes that limit growth at superoptimal temperatures, and squalene epoxidase (ERG1) is predicted to be the most rate limiting. By replacing this single key enzyme with an ortholog from a thermotolerant yeast strain, we obtain a thermotolerant strain that outgrows the wild type, demonstrating the critical role of sterol metabolism in yeast thermosensitivity. Therefore, apart from identifying thermal determinants of cell metabolism and enabling the design of thermotolerant strains, our Bayesian GEM approach facilitates modelling of complex biological systems in the absence of high-quality data and therefore shows promise for becoming a standard tool for genome scale modeling.

Overview

The claims and statistics of this paper haven’t been checked yet.

Free: it reads the open-access text and checks the first 5 claims that cite a source we can check. Create a free account to check every claim.

A few free checks a day · papers already verified open instantly · results are public

Integrity signals

Registry

Crossref · 10.1038/s41467-020-20338-2

Cites

66 of 66 references verified0 citation errors · 0 not checkable

Statistics

Figures

Fig. 1 Using Bayesian statistical learning to integrate temperature dependence in enzyme-constrained GEMs. a An illustration of the temperature effects on enzyme-catalyzed reactions and their integration into an etcGEM (see detailed description and equations in “Methods” section). The metabolic network ecYeast7.616 is shown. b A two-state denaturation model20,21,58 was used to describe the temperature-dependent unfolding process. [E]N is the concentration of the enzyme in the native state; Topt is the optimal temperature at which the specific activity is maximized; Tm and T90 are temperatures at which there is a 50 and 90% probability that an enzyme is in the denatured state, respectively. c Macromolecular rate theory31,33 describing the temperature dependence of enzyme turnover number kcat. Inset shows the heat capacity difference between ground state (E + S) and transition state (E − TS), adapted from Hobbs et al.31. d Temperature dependence of enzyme-specific activity r, which is a product of (b) and (c). e Overview the Bayesian statistical learning approach, where the problem can be formulated as given a generative model (M) (enzyme and temperature constrained genome-scale metabolic model, etcGEM in this study) corresponding to a set of parameters θ and a set of measurements D (phenome data), Bayes’ theorem provides a direct way of updating the Prior distribution of parameters P(θ) to a Posterior distribution P(θ|D): P(θ∣D)=(P(D∣θ)×P(θ))/(P(D)). P(θ|D) is thereby a less uncertain description of the real θ. Since P(D|θ) is, in most applications, computationally expensive or even infeasible to obtain, a sequential Monte Carlo based approximate Bayesian computation (SMC-ABC) approach was implemented (“Methods”) to sample a list of parameter sets from the Posterior.
Fig. 1 Using Bayesian statistical learning to integrate temperature dependence in enzyme-constrained GEMs. a An illustration of the temperature effects on enzyme-catalyzed reactions and their integration into an etcGEM (see detailed description and equations in “Methods” section). The metabolic network ecYeast7.616 is shown. b A two-state denaturation model20,21,58 was used to describe the temperature-dependent unfolding process. [E]N is the concentration of the enzyme in the native state; Topt is the optimal temperature at which the specific activity is maximized; Tm and T90 are temperatures at which there is a 50 and 90% probability that an enzyme is in the denatured state, respectively. c Macromolecular rate theory31,33 describing the temperature dependence of enzyme turnover number kcat. Inset shows the heat capacity difference between ground state (E + S) and transition state (E − TS), adapted from Hobbs et al.31. d Temperature dependence of enzyme-specific activity r, which is a product of (b) and (c). e Overview the Bayesian statistical learning approach, where the problem can be formulated as given a generative model (M) (enzyme and temperature constrained genome-scale metabolic model, etcGEM in this study) corresponding to a set of parameters θ and a set of measurements D (phenome data), Bayes’ theorem provides a direct way of updating the Prior distribution of parameters P(θ) to a Posterior distribution P(θ|D): P(θ∣D)=(P(D∣θ)×P(θ))/(P(D)). P(θ|D) is thereby a less uncertain description of the real θ. Since P(D|θ) is, in most applications, computationally expensive or even infeasible to obtain, a sequential Monte Carlo based approximate Bayesian computation (SMC-ABC) approach was implemented (“Methods”) to sample a list of parameter sets from the Posterior.CC BY · PubMed Central
Fig. 2 Bayesian modeling improves etcGEM performance by reducing parameter statistical uncertainties. a, b Simulated (a) aerobic and (b) anaerobic growth rates in batch cultivations at various temperatures with Prior and Posterior etcGEMs. c Simulated ethanol secretion flux in a chemostat at various temperatures. In a–c, lines indicate median values and shaded areas indicate regions between the 5th and 95th percentiles (n = 128 for Prior and n = 100 for Posterior models). d Principal component analysis (PCA) 21,504 parameter sets (θ^) sampled in the Bayesian approach. Each parameter in the set θ* was standardized by subtracting the mean and then be divided by the standard deviation before PCA. θ^ of 128 Prior and 100 Posterior etcGEMs are highlighted in blue and orange, respectively. All other θ^ were termed as “intermediate” and marked in gray. e The number of enzymes, out of all 764, with a significantly changed mean (Šidák adj. Welch’s t test p value < 0.01, two-sided) and variance (Šidák adj. one-tailed F-test p value < 0.01) in Tm, Topt, and ΔCp^(‡) between Prior and Posterior. Parameters from 128 Prior and 100 Posterior etcGEMs were used for statistical tests. f A random forest model was used to score the importance of all 2292 parameters during the Bayesian approach (“Methods”). The plot shows the accumulated importance score for each of the three-parameter categories. g Parity plot comparing the experimental Tm values and ones in the 100 Posterior models. h Parity plot comparing the experimental Topt values from BRENDA and ones in the 100 Posterior models. In g, h, r denotes the Pearson’s correlation coefficient between experimental values and Posterior mean values. i Comparison among distributions of the mean of Prior Topt s, mean of Posterior Topt s (Post) and experimental Topt s (Exp) from BRENDA. The bimodal distribution of Topt values is due to the common values in BRENDA occurring at room temperature 25 °C (111 times), the optimal growth temperature of yeast 30 °C (244 times), and 37 °C (142 times). Source data are provided as a Source Data file.
Fig. 2 Bayesian modeling improves etcGEM performance by reducing parameter statistical uncertainties. a, b Simulated (a) aerobic and (b) anaerobic growth rates in batch cultivations at various temperatures with Prior and Posterior etcGEMs. c Simulated ethanol secretion flux in a chemostat at various temperatures. In a–c, lines indicate median values and shaded areas indicate regions between the 5th and 95th percentiles (n = 128 for Prior and n = 100 for Posterior models). d Principal component analysis (PCA) 21,504 parameter sets (θ^) sampled in the Bayesian approach. Each parameter in the set θ* was standardized by subtracting the mean and then be divided by the standard deviation before PCA. θ^ of 128 Prior and 100 Posterior etcGEMs are highlighted in blue and orange, respectively. All other θ^ were termed as “intermediate” and marked in gray. e The number of enzymes, out of all 764, with a significantly changed mean (Šidák adj. Welch’s t test p value < 0.01, two-sided) and variance (Šidák adj. one-tailed F-test p value < 0.01) in Tm, Topt, and ΔCp^(‡) between Prior and Posterior. Parameters from 128 Prior and 100 Posterior etcGEMs were used for statistical tests. f A random forest model was used to score the importance of all 2292 parameters during the Bayesian approach (“Methods”). The plot shows the accumulated importance score for each of the three-parameter categories. g Parity plot comparing the experimental Tm values and ones in the 100 Posterior models. h Parity plot comparing the experimental Topt values from BRENDA and ones in the 100 Posterior models. In g, h, r denotes the Pearson’s correlation coefficient between experimental values and Posterior mean values. i Comparison among distributions of the mean of Prior Topt s, mean of Posterior Topt s (Post) and experimental Topt s (Exp) from BRENDA. The bimodal distribution of Topt values is due to the common values in BRENDA occurring at room temperature 25 °C (111 times), the optimal growth temperature of yeast 30 °C (244 times), and 37 °C (142 times). Source data are provided as a Source Data file.CC BY · PubMed Central
Fig. 3 Yeast growth rate is explained by temperature effects on its enzymes. a Illustration of how the temperature dependence of different processes combines to affect the growth rate. Fig. legend: ec—predictions with the enzyme constrained model; ec+NGAM(T)—incorporates the temperature effects on nongrowth associated maintenance into the ec model (Supplementary Fig. 1); ec+kcat(T)—incorporates the temperature effects on enzyme kcat values into the ec model; ec+denaturation(T)—incorporates the temperature effects on enzyme denaturation into the ec model; etc—enzyme and temperature constrained model that includes the temperature effects on NGAM, kcat and enzyme denaturation into ec model. The growth rate at each temperature point was simulated with all 100 Posterior etcGEMs. Lines indicate median values and shaded areas indicate regions between the 5th and 95th percentiles (n = 100). b Comparison between distributions of experimentally measured enzyme Topt values (n = 662) from BRENDA35 and enzyme Tm values (n = 265) from Leuenberger et al.7 in S. cerevisiae. c Comparison between T1/2SA, the temperature at which the specific activity is 50% of its maximum, and T1/2kcat, the temperature at which the kcat value is 50% of its maximum, and Tm in the Posterior models. d Probability of 764 enzymes in the native state. From top to bottom, the enzymes showed increased s. Each pixel represents one probability value of an enzyme at a specific temperature. e Normalized kcat values of 764 enzymes at different temperatures. Each pixel represents one normalized value of an enzyme at a specific temperature. f Normalized specific activities of 764 enzymes at different temperatures. The values in (f) are products of (d) and (e). In d–f, the same ordering of enzymes is shown. Source data are provided as a Source Data file.
Fig. 3 Yeast growth rate is explained by temperature effects on its enzymes. a Illustration of how the temperature dependence of different processes combines to affect the growth rate. Fig. legend: ec—predictions with the enzyme constrained model; ec+NGAM(T)—incorporates the temperature effects on nongrowth associated maintenance into the ec model (Supplementary Fig. 1); ec+kcat(T)—incorporates the temperature effects on enzyme kcat values into the ec model; ec+denaturation(T)—incorporates the temperature effects on enzyme denaturation into the ec model; etc—enzyme and temperature constrained model that includes the temperature effects on NGAM, kcat and enzyme denaturation into ec model. The growth rate at each temperature point was simulated with all 100 Posterior etcGEMs. Lines indicate median values and shaded areas indicate regions between the 5th and 95th percentiles (n = 100). b Comparison between distributions of experimentally measured enzyme Topt values (n = 662) from BRENDA35 and enzyme Tm values (n = 265) from Leuenberger et al.7 in S. cerevisiae. c Comparison between T1/2SA, the temperature at which the specific activity is 50% of its maximum, and T1/2kcat, the temperature at which the kcat value is 50% of its maximum, and Tm in the Posterior models. d Probability of 764 enzymes in the native state. From top to bottom, the enzymes showed increased s. Each pixel represents one probability value of an enzyme at a specific temperature. e Normalized kcat values of 764 enzymes at different temperatures. Each pixel represents one normalized value of an enzyme at a specific temperature. f Normalized specific activities of 764 enzymes at different temperatures. The values in (f) are products of (d) and (e). In d–f, the same ordering of enzymes is shown. Source data are provided as a Source Data file.CC BY · PubMed Central
Fig. 4 Metabolic shifts are explained by temperature-induced proteome constraints. The ATP production in the cytoplasm and the total protein amount required at different temperatures were simulated using Posterior etcGEMs with chemostat culture settings with a dilution rate of 0.1 h−1 (“Methods”). Simulation results from a the original Posterior models, b the original Posterior models with doubled total protein constraint, c the original Posterior models with increased Tm of ATP1, HEM1, and PDB1 (to 50 °C) and are shown. Lines indicate median values and shaded areas indicate the region between the 5th and 95th percentile (n = 100). Source data are provided as a Source Data file.
Fig. 4 Metabolic shifts are explained by temperature-induced proteome constraints. The ATP production in the cytoplasm and the total protein amount required at different temperatures were simulated using Posterior etcGEMs with chemostat culture settings with a dilution rate of 0.1 h−1 (“Methods”). Simulation results from a the original Posterior models, b the original Posterior models with doubled total protein constraint, c the original Posterior models with increased Tm of ATP1, HEM1, and PDB1 (to 50 °C) and are shown. Lines indicate median values and shaded areas indicate the region between the 5th and 95th percentile (n = 100). Source data are provided as a Source Data file.CC BY · PubMed Central
Fig. 5 etcGEM uncovers growth rate-limiting enzymes. a Twenty enzymes with the highest flux sensitivity coefficients at 40 °C (n = 100 for each enzyme). b Predicted maximal specific growth rate of wild-type yeast and the one without any temperature constraints (fully functional) on ERG1 enzyme at 40 °C (n = 100 for each strain). In a, b, centerline, median; box limits, upper and lower quartiles; whiskers, 1.5× interquartile range; Each dot represents the prediction from one Posterior etcGEMs. c The effect of KmERG1 expression on thermotolerance in S. cerevisiae. The strains were passaged for 7 times (24 h for each passage) to obtain stable growth at 40 °C. Optical densities (600 nm) are shown at 24 h. Data are represented as mean values of five replicates in the bar chart. Dots represent the values of five replicates. p Values denote Welch’s t test (two-sided). The exact p values for P1–P7 are 0.307, 0.438, 0.0066, 0.00020, 0.00046, 0.00015, 0.00018, respectively. d Simulated maximum specific growth rate by removing the temperature constraints of most rate-limiting enzymes at each step in each Posterior etcGEM at 42 °C. Lines indicate median values and shaded areas indicate the region between the 5th and 95th percentiles (n = 100). e The percentage of Posterior etcGEMs (n = 100) predicts an enzyme to be in the minimal enzyme set required to be fully functional at 42 °C in order to achieve a maximal specific growth rate of 0.2 h−1. Inset shows the names and pathways of genes predicted by more than Posterior etcGEMs 10% they are involved in. Source data are provided as a Source Data file.
Fig. 5 etcGEM uncovers growth rate-limiting enzymes. a Twenty enzymes with the highest flux sensitivity coefficients at 40 °C (n = 100 for each enzyme). b Predicted maximal specific growth rate of wild-type yeast and the one without any temperature constraints (fully functional) on ERG1 enzyme at 40 °C (n = 100 for each strain). In a, b, centerline, median; box limits, upper and lower quartiles; whiskers, 1.5× interquartile range; Each dot represents the prediction from one Posterior etcGEMs. c The effect of KmERG1 expression on thermotolerance in S. cerevisiae. The strains were passaged for 7 times (24 h for each passage) to obtain stable growth at 40 °C. Optical densities (600 nm) are shown at 24 h. Data are represented as mean values of five replicates in the bar chart. Dots represent the values of five replicates. p Values denote Welch’s t test (two-sided). The exact p values for P1–P7 are 0.307, 0.438, 0.0066, 0.00020, 0.00046, 0.00015, 0.00018, respectively. d Simulated maximum specific growth rate by removing the temperature constraints of most rate-limiting enzymes at each step in each Posterior etcGEM at 42 °C. Lines indicate median values and shaded areas indicate the region between the 5th and 95th percentiles (n = 100). e The percentage of Posterior etcGEMs (n = 100) predicts an enzyme to be in the minimal enzyme set required to be fully functional at 42 °C in order to achieve a maximal specific growth rate of 0.2 h−1. Inset shows the names and pathways of genes predicted by more than Posterior etcGEMs 10% they are involved in. Source data are provided as a Source Data file.CC BY · PubMed Central

Verification

Claims the paper makes

The claims and statistics of this paper haven’t been checked yet.

Free: it reads the open-access text and checks the first 5 claims that cite a source we can check. Create a free account to check every claim.

A few free checks a day · papers already verified open instantly · results are public

Self-check

Percentages, sums and reported differences re-computed from the paper’s own numbers.

The self-check needs the paper’s full text.

Citations

Cites

Show all 66 references

Cited by

See all 57

Metadata

Authors
Li, G., Hu, Y., Zrimec, J., Luo, H., Wang, H., Zelezniak, A., Ji, B., & Nielsen, J.
Affiliations
  • Chalmers University of Technology (SE)
  • Science for Life Laboratory (SE)
  • Novo Nordisk Foundation (DK)
Corresponding author
Nielsen, Jens
Venue
Nature Communications
Published
2021-01-08
Volume
12
Issue
1
Pages
190-190
ISSN
2041-1723 (ISSN-L)
OpenAlex
W3121123436
Cited by
57
References
66
Citations by year
2026: 9 · 2025: 8 · 2024: 11 · 2023: 7 · 2022: 11 · 2021: 10
FWCI
3.06
Type
journal article
Language
en
Open access
Licence
Locations
  • Nature Communications (publishedVersion, cc-by)
  • bioRxiv (Cold Spring Harbor Laboratory) (acceptedVersion, cc-by-nc-nd)
  • Europe PMC (PubMed Central) (publishedVersion, other-oa)
  • PubMed Central (publishedVersion, cc-by)
  • PubMed (publishedVersion)
  • Technical University of Denmark, DTU Orbit (Technical University of Denmark, DTU) (submittedVersion, other-oa)
  • RePEc: Research Papers in Economics (submittedVersion)
  • DOAJ (DOAJ: Directory of Open Access Journals) (submittedVersion, cc-by-sa)
  • Chalmers Research (Chalmers University of Technology) (submittedVersion)
Funders
  • European Commission (HORIZON2020, 722287)
  • Chalmers Tekniska Högskola
  • Knut och Alice Wallenbergs Stiftelse (NNF10CC1016517)
  • Novo Nordisk (NNF10CC1016517)
  • Science for Life Laboratory
  • Novo Nordisk Fonden (NNF17SA0032514, NNF10CC1016517)
Topics
  • Life Sciences › Biochemistry, Genetics and Molecular Biology › Molecular Biology › Microbial Metabolic Engineering and Bioproduction
  • Life Sciences › Biochemistry, Genetics and Molecular Biology › Molecular Biology › Gene Regulatory Network Analysis
  • Life Sciences › Biochemistry, Genetics and Molecular Biology › Molecular Biology › Fungal and yeast genetics research
Keywords
temperature effects on metabolism, yeast metabolism, genome-scale metabolic models, Bayesian parameter estimation, squalene epoxidase, ERG1, sterol metabolism, heat tolerance, growth prediction
MeSH terms
Machine Learning, Bayes Theorem, Models, Biological, Saccharomyces cerevisiae, Temperature, Yeasts, Genome, Fungal, Squalene Monooxygenase

About these checks

Registry, retraction, Cites and statistics results are computed, not judged, with one exception: a printed reference that no index lists is read once by a model before it counts as a citation error. Claims are extracted by a model and kept only when their quote appears verbatim in the text; each cited source is then checked, with a model’s help, for whether it says what the claim relies on. Those verdicts can be wrong — read the claim’s quote and the cited source before relying on one.

Statistics and arithmetic are recomputed, not judged.

Confidence bands are shown only for fields calibrated on our annotated test set. None is calibrated yet, so no band is shown.

What counts as a claim ›How we measure confidence ›How we check this ›

Are you an author, or do you see a problem with this page? Request a review

How to cite

Cite CoChat for the checks on this page, and the paper for its own findings.

This page
CoChat. (2026). Citation check: Bayesian genome scale modelling identifies thermal determinants of yeast metabolism. https://app.cochat.ai/papers/10.1038/s41467-020-20338-2
The paper
Li, G., Hu, Y., Zrimec, J., Luo, H., Wang, H., Zelezniak, A., Ji, B., & Nielsen, J. (2021). Bayesian genome scale modelling identifies thermal determinants of yeast metabolism. Nature Communications. https://doi.org/10.1038/s41467-020-20338-2

Data sources

Bibliographic data: Semantic Scholar, OpenAlex, Crossref.

Contains information from the Semantic Scholar Academic Graph Datasets, made available under the ODC Attribution License (ODC-BY 1.0).

Semantic Scholar’s data is described in: Kinney, R., et al. (2023). The Semantic Scholar Open Data Platform. arXiv:2301.10140.