Improved differential search algorithms for metabolic network optimization
The capabilities of Escherichia coli and Zymomonas mobilis to efficiently converting substrate into valuable metabolites have caught the attention of many industries. However, the production rates of these metabolites are still below the maximum threshold. Over the years, the organism strain design...
Saved in:
| Main Author: | |
|---|---|
| Format: | Thesis |
| Language: | English |
| Published: |
2019
|
| Subjects: | |
| Online Access: | http://eprints.utm.my/id/eprint/96122/1/KautharMohdDaudPSC2019.pdf.pdf http://eprints.utm.my/id/eprint/96122/ http://dms.library.utm.my:8080/vital/access/manager/Repository/vital:143228 |
| Tags: |
Add Tag
No Tags, Be the first to tag this record!
|
| Summary: | The capabilities of Escherichia coli and Zymomonas mobilis to efficiently converting substrate into valuable metabolites have caught the attention of many industries. However, the production rates of these metabolites are still below the maximum threshold. Over the years, the organism strain design was improvised through the development of metabolic network that eases the process of exploiting and manipulating organism to maximize its growth rate and to maximize metabolites production. Due to the complexity of metabolic networks and multiple objectives, it is difficult to identify near-optimal knockout reactions that can maximize both objectives. This research has developed two improved modelling-optimization methods. The first method introduces a Differential Search Algorithm and Flux Balance Analysis (DSAFBA) to identify knockout reactions that maximize the production rate of desired metabolites. The latter method develops a non-dominated searching DSAFBA (ndsDSAFBA) to investigate the trade-off relationship between production rate and its growth rate by identifying knockout reactions that maximize both objectives. These methods were assessed against three metabolic networks – E.coli core model, iAF1260 and iEM439 for production of succinic acid, acetic acid and ethanol. The results revealed that the improved methods are superior to the other state-of-the-art methods in terms of production rate, growth rate and computation time. The study has demonstrated that the two improved modelling-optimization methods could be used to identify near-optimal knockout reactions that maximize production of desired metabolites as well as the organism’s growth rate within a shorter computation time. |
|---|