Hatzimanikatis, VassilyMiskovic, LjubisaTokic, MilenkoHadadi, NoushinAtaman, MeriçEbert, BirgittaBlank, LarsMiskovic, LjubisaHatzimanikatis, Vassily2017-10-032017-10-032017-10-032017https://infoscience.epfl.ch/handle/20.500.14299/140976The limited supply of fossil fuels and environmental reasons shifted research in industry and academia towards finding their replacements. An important part of these efforts is focused on a sustainable production of the 2nd generation of biofuels. Among many proposed molecules, one of the most prominent fuel candidates is Methyl Ethyl Ketone (MEK). MEK, also referred in the literature as 2-butanone, is the most important commercially produced ketone with a broad application as a solvent for paints and adhesives. MEK shows superior characteristics compared the existing fuels regarding thermo-physical and transport properties, increased combustion stability at low engine load and cold boundary conditions while overcoming their limitations on the particle emissions. There is no known native producer of MEK, and there were recent attempts to introduce biosynthetic pathways and produce this molecule in E. coli and yeast. However, none of these attempts resulted in commercially viable amounts of produced MEK. In this study, we used Biochemical Network Integrated Computational Explorer (BNICE.ch) to explore the space of potential biotransformations around MEK. Out of 1325 identified compounds one reaction step away from MEK, we chose 3-oxopentanoic acid, But-3-en-2-one, But-1-en-2-olate and Butylamine for further study as they can: (i) easily be chemically converted to MEK; (ii) be used as precursor metabolites for a range of other valuable chemicals. We have also chosen (2R)-2-Hydroxy-2-methylbutanenitrile for further analysis as it was the only compound with known biotransformation to MEK. We reconstructed 3’679’610 novel biosynthetic pathways up to a length of 4 reaction steps from 157 central carbon metabolites of E. coli toward these 5 compounds. We then embedded the reconstructed pathways in the genome-scale model of the E. coli, and we performed the pathway evaluation with respect to their maximum theoretical yield using first Flux Balance Analysis (FBA) and then Thermodynamic-based Flux Analysis (TFA). Our results showed that thermodynamics is a necessary step in pathway feasibility analysis because a majority of FBA feasible pathways were infeasible in TFA due to the thermodynamic constraints. More precisely, 13.25% (487’411) of all reconstructed pathways were FBA feasible, whereas only 0.51% (18’925) were TFA feasible. We next identified enzymes from the KEGG database with the closest reaction similarity to the novel reactions in the TFA feasible pathways as candidates for direct implementation or enzyme engineering of novel reactions. Furthermore, we classified the TFA feasible pathways depending on the reactions and precursors that were essential for production of the target molecules. Finally, all the feasible pathways were ranked based on the yield, length, number of known reaction steps and enzyme similarity score. In this way, the highest ranked pathways can be considered for experimental implementation. In this study, we illustrated the application of BNICE.ch for discovery and design of novel synthetic pathways and its potential for future developments in the area of metabolic engineering and synthetic biology.Novel reactionsNovel PathwaysMethyl Ethyl KetoneBiofuelsBNICE.ch.text::conference output::conference poster not in proceedings