Abstract:
Experimental Protein-Protein Interaction (PPI) Network data has been collected using high-throughput PPI profiling techniques. PPI network analysis aids in the molecular-level understanding of the proteins. The PPI network alignment can reveal the relationships between multiple species, which improves our knowledge of biological systems and helps to transfer knowledge across the species. PPI network alignment's primary goal is to build a combined network that helps research intricate pathways to identify the roles of unidentified proteins. It aids in the identification of biological processes and molecular-level function understanding of the proteins across species. Through topological and biological similarity, network alignment offers a means of identifying comparable sections across various species and can facilitate the transmission of biological knowledge between them. Several strategies for network alignment have been developed, but it is still difficult to achieve high AFS (Average Functional Similarity) and Coverage. Moreover, the topological methods did not produce quality alignments in terms of average functional similarity. Similarly, the biological information did not guarantee high topological performance. This thesis presents a PPI network alignment algorithm and reviews the existing studies (in terms of semantic similarity and coverage). This thesis investigates different topological measures in addition to biological information to improve the topological and biological performance of the aligners. This thesis presents a novel topological approach that maximizes the AFS, coverage, and overall topological quality.
