Improving the Predictability of Metabolic Network Engineering Design
Lead Research Organisation:
Imperial College London
Department Name: Life Sciences
Abstract
Bioprocess design faces challenges both at the computational and experimental levels. Genome-scale metabolic reconstructions are frequently used to predict optimal pathway yields and growth conditions, which can then be experimentally tested. However, computational simulations do not often resemble what is observed in reality and do not translate into optimal strains in the laboratory. Furthermore, the bottlenecks of experimental implementation span several-fold. Firstly, the identification of the best chemical target is not straightforward. This depends on where within the biotechnological space
there is still freedom to operate, which involves monitoring the latest innovations. A target chemical's physical and chemical properties are also crucial factors, so is the energetic efficiency of the synthetic pathway that produces the target in question. Other concerns include pathway identification, host organism selection, dealing with toxicity, chemical separation and transport. However, the challenge not only relies on the bioprocesses themselves. Market drives, such as supply and demand for specific products, and market prices should also be considered as crucial factors driving the direction of the petrochemicals sector.
To address the above, this PhD project aims to develop new, commercially attractive solutions for the production of petrochemical replacements. Here, the objectives are to develop novel computational tools for more realistic metabolic engineering predictions, and implement whole-chain solutions experimentally, at least at a proof-of-concept level. Further efforts will focus on optimisation and scale-up to increase system efficiency and generate opportunities for commercialisation. This project will strengthen the field's research base by providing new computational and experimental frameworks. It will also attempt to accelerate commercial translation of petrochemical replacements by securing IP and contributing towards business creation, benefiting our overall economic and environmental sustainability.
there is still freedom to operate, which involves monitoring the latest innovations. A target chemical's physical and chemical properties are also crucial factors, so is the energetic efficiency of the synthetic pathway that produces the target in question. Other concerns include pathway identification, host organism selection, dealing with toxicity, chemical separation and transport. However, the challenge not only relies on the bioprocesses themselves. Market drives, such as supply and demand for specific products, and market prices should also be considered as crucial factors driving the direction of the petrochemicals sector.
To address the above, this PhD project aims to develop new, commercially attractive solutions for the production of petrochemical replacements. Here, the objectives are to develop novel computational tools for more realistic metabolic engineering predictions, and implement whole-chain solutions experimentally, at least at a proof-of-concept level. Further efforts will focus on optimisation and scale-up to increase system efficiency and generate opportunities for commercialisation. This project will strengthen the field's research base by providing new computational and experimental frameworks. It will also attempt to accelerate commercial translation of petrochemical replacements by securing IP and contributing towards business creation, benefiting our overall economic and environmental sustainability.
Organisations
People |
ORCID iD |
| Patrik Jones (Primary Supervisor) |
Studentship Projects
| Project Reference | Relationship | Related To | Start | End | Student Name |
|---|---|---|---|---|---|
| BB/M011178/1 | 01/10/2015 | 25/02/2025 | |||
| 1655060 | Studentship | BB/M011178/1 | 01/10/2015 | 30/09/2019 |
| Description | In the growing field of metabolic engineering, microorganisms are treated as potential "factories" to synthesize industrial compounds. The vast number of chemical reactions comprising an organism's metabolism can be harnessed for the production of existing and even novel compounds, including commodity chemicals and fuels. This PhD study focused on exploring how cell metabolism can limit the yield of these commercially-attractive synthetic pathways, and considered the capability of cell metabolism to "mesh" with the energy demands of synthetic pathways, as these demands have the potential to alter the cell's metabolic stability. Insights into what factors render an engineered organism more productive than another are yet to be elucidated. However, we know that from the breakdown of substrate, microorganisms synthesize key co-factors such as ATP and NAD(P)H. For engineered organisms, the balanced supply and consumption of such cofactors, known as cofactor balance, is one such factor influencing their performance. In the present study, we have developed a novel algorithm, coined CBA (Co-factor Balance Assessment), centred on the use of existing computational frameworks, namely FBA, pFBA and FVA. CBA quantifies the co-factor profile of metabolic engineering designs. As a proof-of-concept, various pathways to n-butanol, a non-native product in E. coli, are considered based on how their NADPH and ATP demands are met by the cell. We investigate the network-wide effect of co-factor imbalance in these pathways using an existing computer model of Escherichia coli, the signature microorganism in industrial biotechnology. Our study confirms the excessive flexibility of the FBA framework and evaluates different solutions for this, based on manual and systematic curations. Manual curation of the FBA models prevented co-factor burning and in 87.5% of the cases reached solutions that were theoretically optimal in terms of yield efficiency but were also biomass viable. Systematic constraints reduced the flexibility of FBA but did not circumvent cofactor dissipation in its entirety, although it achieved statistically similar results if these constraints were followed by manual capping. The analysis of existing experimental data also demonstrated that the computational predictions are up to 11.4-fold more flexible than experimentally observed in reality. In a complementary cofactor sensitivity analysis, we predicted an increase in target yield of up to 13% by balancing the introduced engineered pathway both in terms of ATP and redox usage, providing insights into how to balance synthetic pathways to render cell factories more productive. Our methodology was verified using readily available data and compared with existing approaches. We believe that CBA simplifies computational pathway balance assessment, provide new insights into the global network effects of cofactor balancing and facilitate a novel framework to evaluate the cofactor efficiency of recombinant strains. |
| Exploitation Route | The main technical novelty and contribution of this project is a new script for tracking and categorizing how cellular ATP and NADPH are affected (produced or consumed) in the presence of the new pathway. Even though the work presented here is specific to butanol, this code can be adapted for use on any organism and pathway of interest The PhD study focuses specifically on the refinement of this method using manual and systematic curations, and an in-depth evaluation of computational versus experimental predictions based on the system-wide behaviour of existing engineered organisms; However, co-supervision of an additional project at Imperial College London focused on the implementation of this algorithm as a web-based served that non-coders can easily interact with without the need to be tech-savvy. Areas of future work include making the code available as a command-line package. Building up on this and facilitating access across a range of industries would have true impact in the way cell factories are designed, built and evaluated. |
| Sectors | Chemicals,Environment,Manufacturing, including Industrial Biotechology |