Evermore powerful sequencing technologies have made genome sequencing of prokaryotes fast, easy and affordable. However, to unlock the potential for rational engineering or manipulation of strains, genomic information needs to be annotated, and organized in a readily interpretabel way. Metabolic networks can provide such a backbone for the organization and integration of biological data. Indeed, metabolic networks have emerged as a powerful complement to genome annotation by contextualizing the role of individual genes.
The creation of a metabolic network is referred to as network reconstruction, alluding to that the reconstructed network is merely an approximate representation of metabolic network that exist in reality. Canonically, a network reconstruction workflow starts with gene-by-gene genome annotation, followed by successive metabolic reaction additions of reactions catalyzed by proteins encoded by annotated genes. After the initial reconstruction, additional reactions are added to augment the network in such way that it can simulate known functions such as the production of all cellular precursors.
GenomeCraft offers canonical network reconstruction from sequenced genomes, but also offers an integrated annotation-reconstruction process, were the mutually corroborating information of the emerging network structure is used for simultaneous annotation and reconstruction. GnomeCraft offers an array of network reconciliation options, including minimal biomass and maximum precursor production solutions.
GenomeCraft reconciles networks using weighted linear programming techniques that were developed in the Libourel group at the University of Minnesota. In addition to thermodynamically informed reversibility constraints and sequence similarity-based reaction addition, weighted linear programming is exceptionally suitable for the reconstruction of multi-compartment networks such as for Eukaryotes.
Relevant publications: