Exploring Symmetric Substructures in Protein Interaction Networks for Pairwise Alignment

In molecular biology, comparison of multiple Protein Protein Interaction (PPI) networks to extract subnetworks that are conserved during evolution across di↵erent species is helpful for studying complex cellular machinery. Most e↵orts produce promising results in creating alignments that show large regions of biological or topological similarity between the PPI networks of various species, but few do both. We present a new pairwise aligner SSAlign (Symmetric Substructure Alignment) that extracts maximal substructures from participating PPI networks and uses Gene Ontology Consistency (GOC) as the graph isomorphic function for aligning two subgraphs. We use PPI networks from Isobase data repository for experiments and comparisons. Our results show that in comparison to other contemporary aligners, SSAlign is better at aligning topologically and biologically similar subnetworks.