The complexity of the metabolic pathway interaction leads to the accepted truth that it is impossible to separately analyze a single pathway and proteomic information has shown that genetic modifications pertaining a specific pathway also affect other metabolic processes. Genome-scale models have the whole biochemical network in perspective and can predict the effects of genetic modifications in microbial metabolism, such as uptake/secretion rates, exceedingly better than single pathway analysis. Constructing genome scale models is an extremely laborious task. Also, in every microorganism the genome is not yet fully annotated or has errors which will require some degree of manual correction. In our work, we suggest that the development time for metabolic model building can be decreased with usage of automated import procedures from public databases such as KEGG and by eliminating or correcting, a priori, biochemical reactions that will result in metabolic dead ends. These consist in specific compounds which lack known metabolic routes that prevent them from accumulating in the cell. Our algorithms have imported and reconstructed the networks from 619 microbes available in KEGG as well as a hyper network consisting of every known metabolic reaction (7353 in total). Using the hyper network, we also successfully identified 2817 reactions (38%) responsible for 2315 general metabolic dead ends. We have shown that metabolic networks and proteomic gel analysis can be mutually enhanced. This combination provides us with additional model constraints and increases the gel resolution. For this project, we reconstructed the network of Aspergillus niger and analyzed the proteomic gels of the wild strain and single gene mutant.