Page 113 - The-5th-MCAIT2021-eProceeding

Improving Production Rate and Growth Rate of Mutants: A
Comparison of Constraint-Based Modeling Approaches

a*
c
b
Kauthar Mohd Daud , Zalmiyah Zakaria , Zuraini Ali Shah
a Center for Artificial Intelligence Technology, Faculty of Information Science and Technology, Universiti Kebangsaan Malaysia, 43600
Bangi, Selangor, Malaysia
b, c Artificial Intelligence and Bioinformatics (AIBIG), School of Computing, Faculty of Engineering, Universiti Teknologi Malaysia,
81310 Johor Bahru, Johor, Malaysia
* Email: kauthar.md@ukm.edu.my

Abstract
In silico metabolic engineering is a process to improve the phenotypic characteristics by manipulating the genotypic
elements of an organism. Reaction knockout is one of the genetic perturbation strategies use to analyse the effects in
optimizing the production rate. The optimality of an organism can be predicted and simulated using constraint-based
modeling (CBM) approaches. In this study, the stoichiometric model of Escherichia coli was examined to improve the
production of succinic acid. Three well-known CBM approaches, which are Flux Balance Analysis (FBA), Minimization
of Metabolic Adjustment (MoMA) and Regulatory On/Off Minimization (ROOM), were compared. The comparison
resulted in a higher production rate predicted by MoMA and higher growth rate by FBA. Regardless, FBA is more stable in
simulating the reaction knockout as it predicts the final steady-state of organisms after the genetic perturbation.

Keywords: constraint-based modeling method; reaction knockout; genome-scale metabolic network; in silico metabolic engineering

1. Introduction

In the early 1970s, the chemical industry has started to change into modern biotechnologies to fulfil the
increasing demands of valuable products, for instance, pharmaceuticals, food ingredients and bio-based fuel
(Maia, Rocha and Rocha, 2016). Advancements in genome sequencing have allow researchers to have insight
on an organism. Therefore, the term “metabolic engineering” has coined that allow researchers to probe in detail
the biological elements of an organism, thus gives full capability to exploit the genetic capability of an organism
for strains improvement. The aims of metabolic engineering are to optimize the metabolism of organisms by
exploiting and manipulating their metabolic capabilities through modelling. Thus, generates economically and
industrially viable organisms through optimization and predictive tools.
In silico metabolic engineering is a scientific domain that integrates with the computational technology to
allows faster simulation and design predictions of a mutant. It entails reconstructing the mathematical
representation of the metabolic network, investigating the effects of genetic perturbations and their relations to
the phenotype characteristics. Consequent to that, the development and reconstruction of genome-scale
metabolic networks have rapidly grown and to keep up with the huge data, more sophisticated methods and
algorithms are needed.
The approaches in metabolic engineering can be divided into two, which are the dynamic approach and static
approach that varies in metabolic representation whereby dynamic approach uses kinetic modelling and static
approach uses a stoichiometric matrix to represent the metabolic network (Suthers et al., 2021). Stoichiometric
models are commonly used as modelling frameworks compared to kinetic models as it does not requires
difficult-to-obtained experimental data for parameter estimation (Vasilakou et al., 2016). In stoichiometric
models, the biochemical reactions in the metabolic network are represented as a set of stoichiometric equations,
whereby the elements of different metabolites in the metabolic network are denoted as stoichiometric
coefficients in the stoichiometric matrix.

E- Proceedings of The 5th International Multi-Conference on Artificial Intelligence Technology (MCAIT 2021) [100]
Artificial Intelligence in the 4th Industrial Revolution