Path to improving the life cycle and quality of genome-scale models of metabolism

Abstract

Genome-scale models of metabolism (GEMs) are key computational tools for the systems-level study of metabolic networks. Here, we describe the ''GEM life cycle,'' which we subdivide into four stages: inception, maturation, specialization, and amalgamation. We show how different types of GEM reconstruction workflows fit in each stage and proceed to highlight two fundamental bottlenecks for GEM quality improvement: GEM maturation and content removal. We identify common characteristics contributing to increasing quality of maturing GEMs drawing from past independent GEM maturation efforts. We then shed some muchneeded light on the latent and unrecognized but pervasive issue of content removal, demonstrating the substantial effects of model pruning on its solution space. Finally, we propose a novel framework for content removal and associated confidence-level assignment which will help guide future GEM development efforts, reduce duplication of effort across groups, potentially aid automated reconstruction platforms, and boost the reproducibility of model development. ll

Systems-level patterns in biological processes are changed under prolongevity interventions and across biological age

“…As a complementary approach to the findings from DIRAC and WGCNA analyses, we performed in silico analysis using the mouse GEM 43 to investigate metabolic shifts associated with prolongevity interventions. GEM is a mathematical framework that leverages knowledge-base cataloging information about biochemical reactions within a system (e.g., single cell, tissue, organ), including metabolites, genes encoding catalytic enzymes, and their stoichiometry 44 . Using optimization techniques with large-scale experimental data (e.g., transcriptomics), the solved stoichiometric coefficients of each reaction allow flux prediction for metabolic reactions in the system at equilibrium 53 .…”

“…We apply DIRAC analysis to mouse liver protein abundance profiles, first with predefined modules derived from Gene Ontology Biological Process (GOBP) annotations and then with unbiased modules derived from Weighted Gene Co-expression Network Analysis (WGCNA) 41,42 , and demonstrate that three lifespan-extending drugs (ACA, 17aE2, and Rapa) promoted tighter regulation of aging-related modules, such as fatty acid metabolism and inflammation processes. As a complementary approach, mouse genome-scale metabolic model (GEM) 43,44 is developed with the three drugs-including liver transcriptomics 45 , and exhibits that multiple prolongevity interventions shifted fatty acid metabolism. In addition, comparisons of DIRAC analyses between the liver proteomics and transcriptomics suggest that biological modules were tightly regulated by the prolongevity interventions at different levels: transcription vs. post-transcription including the cap-independent translation (CIT) of specific mRNAs 46 .…”

Genome-scale modeling of yeast metabolism: retrospectives and perspectives

“… 2018 ; Zuñiga, Tibocha-Bonilla and Betenbaugh 2021 ). In contrast to the addition efforts, removal of redundant entries and low confidence content was recently highlighted, which has already been observed during the update of yeast GEMs and is believed to be more efficiently performed in yeast GEM refinement with the proposed content removal framework (Seif and Palsson 2021 ). Another improvement could be on the biomass equation, which has been almost condition-independent and scarcely changed in yeast GEM sequels although Yeast8 has formulated metal ions and vitamins of the biomass composition.…”

Reconstructing Kinetic Models for Dynamical Studies of Metabolism using Generative Adversarial Networks

“…Recent technological progress in high-throughput measurement techniques has propelled discoveries in biology, biotechnology, and medicine, and allowed us to integrate multiple different data types into representations of cellular states and obtain deep insights into cellular physiology. Historically, researchers have used genome-scale models (mathematical descriptions of cellular metabolism), to associate experimentally observed biological data with cellular phenotype [1][2][3] . However, traditional genome-scale models cannot predict the dynamic cellular responses to internal or external stimuli because they lack information about metabolic regulation and enzyme kinetics 4,5 .…”