Background: Various properties of Protein-Protein Interaction (PPI) network have been
widely exploited to discover the topological organizing principle and the crucial function motifs
involving specific biological pathway or disease process. The current motifs of PPI network are
either detected by the topology-based coarse grain algorithms, i.e. community discovering, or
depended on the limited-accessible protein annotation data derived precise algorithms. However,
the identified network motifs are hardly compatible with the well-defined biological functions
according to those two types of methods.
Method: In this paper, we proposed a minimal protein loop finding method to explore the
elementary structural motifs of human PPI network. Initially, an improved article exchange model
was designed to search all the independent shortest protein loops of PPI network. Furthermore,
Gene Ontology (GO) based function clustering analysis was implemented to identify the biological
functions of the shortest protein loops. Additionally, the disease process associated shortest protein
loops were considered as the potential drug targets.
Result: Our proposed method presents the lowest computational complexity and the highest
functional consistency, compared to the three other methods. The functional enrichment and
clustering analysis for the identified minimal protein loops revealed the high correlation between
the protein loops and the corresponding biological functions, particularly, statistical analysis
presenting the protein loops with the length less than 4 is closely connected with some disease
process, suggesting the potential drug target.
Conclusion: Our minimal protein loop method provides a novel manner to precisely define the
functional motif of PPI network, which extends the current knowledge about the cooperating
mechanisms and topological properties of protein modules composed of the short loops.