You're using an outdated browser. Please upgrade to a modern browser for the best experience.
Multiomics-Integrated Genome-scale Metabolic Models: Comparison
View Latest Version
Please note this is a comparison between Version 3 by Nora Tang and Version 2 by Nora Tang.
Although the multiconstraint approach in GEMs allows researchers to explore cellular metabolic networks, there are still certain difficulties in analyzing complex regulatory mechanisms in cells. Therefore, GEMs integrating the transcriptional regulatory network (TRN) and protein structure (PRO) were constructed to comprehensively analyze the regulatory mechanism of the metabolic network in cells and the feedback regulation of metabolic flux at the genome scale to understand the growth and metabolic process of cells in detail.
 
  • multiscale genome-scale metabolic models
  • multiconstraint models
  • multiomics models

1. TRN-Integrated GEMs

Transcriptional regulation is one of the important mechanisms by which microorganisms transform their metabolic flux in response to changing environments. TRNs have been widely reported in bacteria after the standard procedure for reconstituting TRNs was proposed [1]. TRNs usually appear as a network of mutual regulation between genes, and global transcription factors control the expression of most genes.
Two tool platforms for the integration of TRN into GEMs based on logical Boolean rules were developed: the toolbox for integrating genome-scale metabolism (TIGER) [2] and FlexFlux [3]. TIGER converts generalized Boolean and multilevel rules into MILPs and couples these rules into GEMs to address the multiple iterations required to reach a steady-state for multilayered transcriptional regulation compared to traditional single iterations [2]. Unlike TIGER, FlexFlux has a user-friendly graphical interface, and it applies the regulatory steady-state analysis algorithm to constrain each component in the network to a single steady-state [3]. Importantly, FlexFlux allows the transformation of discrete qualitative states of regulatory networks into user-defined continuous intervals and the different approaches to a detailed analysis of regulatory mechanisms in metabolic network models [3].
Furthermore, the probabilistic regulation of metabolism (PROM) [4], gene expression and metabolism integrated for network inference [5], and transcriptional regulation FBA [6] realized the coupling of GEMs and transcriptional regulation models based on Boolean rules and explored the effects of transcription factors on the cell phenotype in different environments. Based on PROM, the integrated deduced and metabolism (IDREAM) method [7] and the optimization of regulatory and metabolic networks approach (OptRAM) [8] were developed to evaluate the regulatory role of transcription factors in metabolic networks. The strategy of optimal gene combination optimization can be inferred to improve the yield of the target product.
For E. coli, TRN-integrated GEMs were constructed from quantitative cell growth data [4]. Six strains with key transcriptional regulator knockout in the oxygen consumption reactions were constructed according to the model prediction. The model accurately predicted the growth rate of 14 knockout phenotypes, with a correlation coefficient of 0.95 [4]. For Mycobacterium tuberculosis, an expanded knowledge base of metabolic networks and regulatory mechanisms was constructed with 104 TF regulatory networks based on ChIP-seq interactions linked to 810 GEMs [9]. The knowledge base identified synergistic TF–drug interactions in >50% of the cases, suggesting that this model may provide corresponding information for antituberculosis drug target identification [9]. For S. cerevisiae, a TRN-integrated GEM was constructed involving 25,000 regulatory interactions and controlling 1597 metabolic reactions [5]. The model accurately predicted the phenotype of TF knockout under different conditions and revealed potential condition-specific regulatory mechanisms [5]. Furthermore, Shen et al. used OptRAM to design efficient synthetic strains of succinic acid, 2,3-butanediol, and ethanol in yeast and confirmed the role of key predicted genes [8]. The productivity of 2,3-butanediol increased by 61 times compared to the experimental value under the optimization strategy simulation, and the productivity of ethanol increased by 1.8 times under the same conditions [8].

2. PRO-Integrated GEMs

GEM construction relies on the mining of multiomics and the analysis of cellular metabolic processes, in which protein–protein interactions control a wide range of cellular processes, such as signal transduction [9][10] and molecular transport [11]. Therefore, introducing proteomic data into GEMs can provide insights into metabolic network mechanisms in cells [12].
Brunk et al. proposed the GEM with a protein structure (GEM-PRO) framework, which directly maps genes to transcripts, PROs, biochemical responses, network states, and, ultimately, phenotypes [13]. The massive open-source protein database provides >110,000 entries of biological macromolecular structure information [14]. These have facilitated the development of protein ensemble models. Chang et al. integrated GEMs with data such as amino acid sequence, PRO, functional annotation, and protein-substrate binding sites to analyze protein stability in the cellular environment [15]. PRO-integrated GEMs predicted the growth-limiting factor of heat resistance and revealed the metabolic mechanism of heat resistance for E. coli [15]. GEM-PROs of E. coli and T. maritima were reported, revealing growth limitation by protein instability through features such as temperature conditions, protein folding, and substrate specificity [13]. The establishment of this model demonstrates the utility of the intersection of systems biology and structural biology [13].
Recently, an integrated GEM based on protein synthesis and degradation was reported in yeast, which systematically alters the growth rate and determines its protein expression level [16]. Importantly, this model identifies protein compartment-specific constraints to reveal growth rate-optimized protein expression profiles, providing a framework for understanding metabolic mechanisms in eukaryotic cells [16]. However, except for E. coli and yeast, PRO-integrated GEMs have not been widely used, and the acquisition of accurate PRO data may be the main limiting factor for its development.

3. Comprehensive Metabolic Models

Cellular metabolism is regulated at multiple levels, so a single integrated model cannot accurately predict cellular phenotypes under various environmental conditions. Therefore, the development of comprehensive models facilitates the exploration of cellular metabolism at multiscale levels.
In E. coli, a comprehensive modeling framework (EcoMAC), which unifies various biological processes and multilayer interactions, was developed to combine gene expression data from genetic and environmental perturbations, transcriptional regulation, signal transduction and metabolic pathways, and growth measurements [17]. In this framework, expression balance analysis was used to integrate genetic, competence, phenomenological, and environmental constraints to predict gene expression, and a new approach to transcription-based metabolic flux enrichment was developed to expand flux boundaries and simultaneously calculate metabolic interaction with transcription [17]. Notably, EcoMAC improved the performance of the region classifier to 22%, identifying stress responses, locomotion and taxis, and cell motility, which were the most abundant biological processes from 500 computationally inferred interactions [17]. A knowledge base calculating the traits of E. coli, iML1515, was reported, which contained not only transcriptome, proteome, and metabolome data, but also unique metabolite response information and complete PRO data [18]. The knowledge base simulated 23,617 phenotypic data with 93.4% accuracy in gene knockouts of 16 different carbon sources and identified important metabolic differences in clinical isolates [18]. These all reflect its potential for identifying drug targets and then using them in therapeutic and clinical applications.
For S. cerevisiae, a genome-wide tool for multiscale modeling data extraction and representation (GEMMER) was developed. This tool aids the visualization of the physical, regulatory, and genetic interactions between proteins and genes and integrates existing database information to support multiscale modeling efforts [19]. Lu et al. introduced a model ecosystem based on the Yeast8 model platform, which includes ecYeast8 (enzyme constraint model), panYeast8 (protein 3D structure database), and coreYeast8 (core metabolic network model of 1011 different mutant strains of S. cerevisiae) [20]. This model ecosystem comprehensively explores the effects of single nucleotide variation on phenotypic characteristics, promotes the exploration of yeast metabolism at the multiscale level, and provides guidance for the wide application of yeast systems and synthetic biology [20].

References

  1. Faria, J.; Overbeek, R.; Xia, F.; Rocha, M.; Rocha, I.; Henry, C.S. Genome-scale bacterial transcriptional regulatory networks: Reconstruction and integrated analysis with metabolic models. Briefings Bioinform. 2013, 15, 592–611.
  2. Jensen, P.A.; Lutz, K.A.; Papin, J.A. TIGER: Toolbox for integrating genome-scale metabolic models, expression data, and transcriptional regulatory networks. BMC Syst. Biol. 2011, 5, 147.
  3. Marmiesse, L.; Peyraud, R.; Cottret, L. FlexFlux: Combining metabolic flux and regulatory network analyses. BMC Syst. Biol. 2015, 9, 93.
  4. Chandrasekaran, S.; Price, N.D. Probabilistic integrative modeling of genome-scale metabolic and regulatory networks in Escherichia coli and Mycobacterium tuberculosis. Proc. Natl. Acad. Sci. USA 2010, 107, 17845–17850.
  5. Chandrasekaran, S.; Price, N.D. Metabolic Constraint-Based Refinement of Transcriptional Regulatory Networks. PLOS Comput. Biol. 2013, 9, e1003370.
  6. Motamedian, E.; Mohammadi, M.; Shojaosadati, S.A.; Heydari, M. TRFBA: An algorithm to integrate genome-scale metabolic and transcriptional regulatory networks with incorporation of expression data. Bioinformatics 2017, 33, 1057–1063.
  7. Wang, Z.; Danziger, S.A.; Heavner, B.D.; Ma, S.; Smith, J.J.; Li, S.; Herricks, T.; Simeonidis, E.; Baliga, N.S.; Aitchison, J.D.; et al. Combining inferred regulatory and reconstructed metabolic networks enhances phenotype prediction in yeast. PLOS Comput. Biol. 2017, 13, e1005489.
  8. Shen, F.; Sun, R.; Yao, J.; Li, J.; Liu, Q.; Price, N.D.; Liu, C.; Wang, Z. OptRAM: In-silico strain design via integrative regulatory-metabolic network modeling. PLOS Comput. Biol. 2019, 15, e1006835.
  9. Ma, S.; Minch, K.J.; Rustad, T.R.; Hobbs, S.; Zhou, S.-L.; Sherman, D.R.; Price, N.D. Integrated Modeling of Gene Regulatory and Metabolic Networks in Mycobacterium tuberculosis. PLOS Comput. Biol. 2015, 11, e1004543.
  10. Friedman, A.; Perrimon, N. Genetic Screening for Signal Transduction in the Era of Network Biology. Cell 2007, 128, 225–231.
  11. Nyfeler, B.; Michnick, S.W.; Hauri, H.-P. Capturing protein interactions in the secretory pathway of living cells. Proc. Natl. Acad. Sci. USA 2005, 102, 6350–6355.
  12. Chen, Y.; Nielsen, J. Mathematical modeling of proteome constraints within metabolism. Curr. Opin. Syst. Biol. 2021, 25, 50–56.
  13. Brunk, E.; Mih, N.; Monk, J.; Zhang, Z.; O’Brien, E.J.; Bliven, S.E.; Chen, K.; Chang, R.L.; Bourne, P.E.; Palsson, B.O. Systems biology of the structural proteome. BMC Syst. Biol. 2016, 10, 26.
  14. Rose, P.W.; Bi, C.; Bluhm, W.F.; Christie, C.H.; Dimitropoulos, D.; Dutta, S.; Green, R.K.; Goodsell, D.S.; Prlic, A.; Quesada, M.; et al. The RCSB Protein Data Bank: New resources for research and education. Nucleic Acids Res. 2012, 41, D475–D482.
  15. Chang, R.L.; Andrews, K.; Kim, D.; Li, Z.; Godzik, A.; Palsson, B.O. Structural Systems Biology Evaluation of Metabolic Thermotolerance in Escherichia coli. Science 2013, 340, 1220–1223.
  16. Elsemman, I.E.; Prado, A.R.; Grigaitis, P.; Albornoz, M.G.; Harman, V.; Holman, S.W.; van Heerden, J.; Bruggeman, F.J.; Bisschops, M.M.M.; Sonnenschein, N.; et al. Whole-cell modeling in yeast predicts compartment-specific proteome constraints that drive metabolic strategies. Nat. Commun. 2022, 13, 801.
  17. Carrera, J.; Estrela, R.; Luo, J.; Rai, N.; Tsoukalas, A.; Tagkopoulos, I. An integrative, multi-scale, genome-wide model reveals the phenotypic landscape of Escherichia coli. Mol. Syst. Biol. 2014, 10, 735.
  18. Monk, J.M.; Lloyd, C.J.; Brunk, E.; Mih, N.; Sastry, A.; King, Z.; Takeuchi, R.; Nomura, W.; Zhang, Z.; Mori, H.; et al. iML1515, a knowledgebase that computes Escherichia coli traits. Nat. Biotechnol. 2017, 35, 904–908.
  19. Mondeel, T.D.G.A.; Cremazy, F.; Barberis, M. GEMMER: GEnome-wide tool for Multi-scale Modeling data Extraction and Representation for Saccharomyces cerevisiae. Bioinformatics 2018, 34, 2147–2149.
  20. Lu, H.; Li, F.; Sánchez, B.; Zhu, Z.; Li, G.; Domenzain, I.; Marcišauskas, S.; Anton, P.M.; Lappa, D.; Lieven, C.; et al. A consensus S. cerevisiae metabolic model Yeast8 and its ecosystem for comprehensively probing cellular metabolism. Nat. Commun. 2019, 10, 3586.
More
© Text is available under the terms and conditions of the Creative Commons Attribution (CC BY) license (http://creativecommons.org/licenses/by/4.0/)
Academic Video Service
Feedback