PARROT: Prediction of enzyme abundances using protein-constrained metabolic models

Abstract

Motivation: Protein allocation determines activity of cellular pathways and affects growth across all organisms. Therefore, a variety of experimental and machine learning approaches has been developed to quantify and predict protein abundances, respectively. Yet, despite advances in protein quantification, it remains challenging to predict condition-specific allocation of enzymes in metabolic networks. Results: Here we propose a family of constrained-based approaches, termed PARROT, to predict enzyme allocations based on the principle of minimizing the enzyme allocation adjustment using protein constrained metabolic models. To this end, PARROT variants model the minimization of enzyme reallocation using four different (combinations of) distance functions. We demonstrate that the PARROT variant that minimizes the Manhattan distance of enzyme allocations outperforms existing approaches based on the parsimonious distribution of fluxes or enzymes for both Escherichia coli and Saccharomyces cerevisiae. Further, we show that the combined minimization of flux and enzyme allocation adjustment leads to poor and inconsistent predictions. Together, our findings indicate that minimization of resource rather than flux redistribution is a governing principle determining steady-state pathway activity for microorganism grown in suboptimal conditions. Availability and implementation: The implementation of PARROT can be found in the GitHub repository: https://github.com/mauricioamf/PARROT

Protein constraints in genome‐scale metabolic models: Data integration, parameter estimation, and prediction of metabolic phenotypes

“…Two approaches based on the minimization of metabolic adjustment principles have been developed to specifically predict the adjustment of enzyme usage. The first approach, termed PARROT (Ferreira et al, 2023), proposes the minimization of the Manhattan (linear) or Euclidean (quadratic) distances between the enzyme usage distribution of a reference growth condition and an alternative growth condition, with or without the consideration of metabolic fluxes. PARROT is available on a GitHub repository, and is implemented in MATLAB, as is therefore compatible with pcGEMs generated with the GECKO Toolbox (Sánchez et al, 2017).…”

“…Data‐driven models could be used to overcome this problem, as features used to train the models such as mRNA expression and codon usage bias can be a proxy for in vivo enzyme concentrations. A coupling of data‐driven and constraint‐based models could further enhance predictions, as proposed in the CAMEL approach (Ferreira et al, 2023), where the ratio between pcGEM‐predicted protein abundance values and in vivo measurements was used to train machine learning models. This development highlights the opportunities for integrating multiomics and multimodel approaches.…”

Data integration across conditions improves turnover number estimates and metabolic predictions

“…Since PRESTO considers protein abundances for the correction, which is not the case for GECKO, we expected to find the increased prediction performance with PRESTO compared to GECKO; however, we still observe low overall predictability of protein abundances using the resulting models. Recently, a more sophisticated protein abundance prediction approach using pcGEMs was introduced that can be used to predict more reliable values and might further be improved by considering corrections introduced by PRESTO 30 .…”