A Novel Reconstruction Pipeline for Genome-Based Network Models Using High Quality Reference Networks

The rapid drop in sequencing cost has enabled companies to include re-sequencing of production strain candidates as a routine analytical tool in strain development projects. Moreover, easy availability of genome sequences and genome annotations also fosters increased acceptance of novel, less‑characterized organisms as production hosts for specialty products. Several automated pipelines have recently been developed that facilitate the automated annotation of new genomes and some even support establishment of associated network reconstructions. Nevertheless, obtaining high-quality network models that permit performing functional assessment of novel strains and strain variants remains challenging.

Here, we present an automated reconstruction pipeline that takes the annotated genome sequence of the new strain of interest and high-quality network models of reference organisms as input to rapidly produce a network reconstruction for the new strain. Choosing high‑quality reference models that have proven simulation capabilities and that have been verified with experimental data greatly facilitates obtaining simulation-ready reconstructions and identification of functional gaps.

The new reconstruction pipeline is integrated with the Insilico Discovery™ software and capitalizes on its functionalities for graphical visualization of the reconstructed network, gap identification, and assessment of network functionality. Graphical comparison with a reference network serves to quickly identify metabolic differences, e.g., between successive strain variants or between different production hosts. Moreover, the expected impact on the metabolic phenotype can be quantified using the extensive simulation capabilities of the Insilico Discovery™ software, ranging from constrained-based analysis methods to dynamic simulation of the resulting network in a bioprocess environment.

In this way, the novel reconstruction pipeline enables rapid analysis of genotype-phenotype relationships based on re-sequencing data in the field of rational strain development.