TY - JOUR
T1 - Strain design optimization using reinforcement learning
AU - Sabzevari, Maryam
AU - Szedmak, Sandor
AU - Penttilä, Merja
AU - Jouhten, Paula
AU - Rousu, Juho
N1 - Funding Information:
JR was supported by grants from The Finnish Innovation Fund SITRA (www.sitra.fi), grant number 381202 and Academy of Finland (www.aka.fi) grant number 310107. PJ was supported by grants from Academy of Finland (https://www.aka.fi/en/) with grant numbers 310514 and 314125. MP was supported by grant from Jenny and Antti Wihuri Foundation (wihurinrahasto.fi). The funders had no role in study design, data collection and analysis, decision to publish, or preparation of the manuscript.
Publisher Copyright:
© 2022 Sabzevari et al. This is an open access article distributed under the terms of the Creative Commons Attribution License, which permits unrestricted use, distribution, and reproduction in any medium, provided the original author and source are credited.
PY - 2022/6
Y1 - 2022/6
N2 - Engineered microbial cells present a sustainable alternative to fossil-based synthesis of chemicals and fuels. Cellular synthesis routes are readily assembled and introduced into microbial strains using state-of-the-art synthetic biology tools. However, the optimization of the strains required to reach industrially feasible production levels is far less efficient. It typically relies on trial-and-error leading into high uncertainty in total duration and cost. New techniques that can cope with the complexity and limited mechanistic knowledge of the cellular regulation are called for guiding the strain optimization. In this paper, we put forward a multi-agent reinforcement learning (MARL) approach that learns from experiments to tune the metabolic enzyme levels so that the production is improved. Our method is model-free and does not assume prior knowledge of the microbe’s metabolic network or its regulation. The multi-agent approach is well-suited to make use of parallel experiments such as multi-well plates commonly used for screening microbial strains. We demonstrate the method’s capabilities using the genome-scale kinetic model of Escherichia coli, k-ecoli457, as a surrogate for an in vivo cell behaviour in cultivation experiments. We investigate the method’s performance relevant for practical applicability in strain engineering i.e. the speed of convergence towards the optimum response, noise tolerance, and the statistical stability of the solutions found. We further evaluate the proposed MARL approach in improving L-tryptophan production by yeast Saccharomyces cerevisiae, using publicly available experimental data on the performance of a combinatorial strain library. Overall, our results show that multi-agent reinforcement learning is a promising approach for guiding the strain optimization beyond mechanistic knowledge, with the goal of faster and more reliably obtaining industrially attractive production levels.
AB - Engineered microbial cells present a sustainable alternative to fossil-based synthesis of chemicals and fuels. Cellular synthesis routes are readily assembled and introduced into microbial strains using state-of-the-art synthetic biology tools. However, the optimization of the strains required to reach industrially feasible production levels is far less efficient. It typically relies on trial-and-error leading into high uncertainty in total duration and cost. New techniques that can cope with the complexity and limited mechanistic knowledge of the cellular regulation are called for guiding the strain optimization. In this paper, we put forward a multi-agent reinforcement learning (MARL) approach that learns from experiments to tune the metabolic enzyme levels so that the production is improved. Our method is model-free and does not assume prior knowledge of the microbe’s metabolic network or its regulation. The multi-agent approach is well-suited to make use of parallel experiments such as multi-well plates commonly used for screening microbial strains. We demonstrate the method’s capabilities using the genome-scale kinetic model of Escherichia coli, k-ecoli457, as a surrogate for an in vivo cell behaviour in cultivation experiments. We investigate the method’s performance relevant for practical applicability in strain engineering i.e. the speed of convergence towards the optimum response, noise tolerance, and the statistical stability of the solutions found. We further evaluate the proposed MARL approach in improving L-tryptophan production by yeast Saccharomyces cerevisiae, using publicly available experimental data on the performance of a combinatorial strain library. Overall, our results show that multi-agent reinforcement learning is a promising approach for guiding the strain optimization beyond mechanistic knowledge, with the goal of faster and more reliably obtaining industrially attractive production levels.
KW - Escherichia coli/genetics
KW - Metabolic Engineering/methods
KW - Metabolic Networks and Pathways
KW - Saccharomyces cerevisiae/genetics
KW - Synthetic Biology
UR - http://www.scopus.com/inward/record.url?scp=85131791539&partnerID=8YFLogxK
U2 - 10.1371/journal.pcbi.1010177
DO - 10.1371/journal.pcbi.1010177
M3 - Article
C2 - 35658018
AN - SCOPUS:85131791539
VL - 18
JO - PLoS Computational Biology
JF - PLoS Computational Biology
SN - 1553-734X
IS - 6
M1 - e1010177
ER -