Supplementary resources to: “Using the Gene Ontology tool to produce de novo protein-protein interaction networks with IS_A relationship”
In the file “Gene_Ontology_de_novo_PPI.zip”, I present data extracted from the database used to coin the article " Using the Gene Ontology tool to produce de novo protein-protein interaction networks with IS_A relationship". There are protein-protein interaction (PPI) networks for all the ten organisms mentioned in the article, besides their respective plasmids. However, the PPIs available for download differ from those published since I didn't restringed them only to true positives according to the String database. Instead of that, I considered candidate relationships all protein pairs possessing commonality between all the three Gene Ontology categories. The edges weight reflects a logarithmic distance between protein pairs measured over the gene position within a chromosome. According to the methodology applied in the paper, a pair of genes separated by five loci has the weight=(1+(MAX-log(pos(locus_00006)-pos(locus_00001)))). In this formulae, pos extracts the locus_tag index; the logarithm of the difference is summed to one to avoid edges smaller than one because it could not be accepted by some visualization tools like GEPHI; MAX creates thicker lines for closer pairs. Figure 1 depicts data from the file "Escherichia_coli_S88_p1.dot".