Association Study based on Topological Constraints of Protein-Protein Interaction Networks

Abstract

The non-random interaction pattern of a protein-protein interaction network (PIN) is biologically informative but its potentials have not been fully utilized in omics studies. Here, we propose a network-permutation-based association study (NetPAS) method that gauges the observed interactions between two sets of genes based on the comparison between permutation null models and the empirical networks. This enables NetPAS to evaluate relationships, constrained by network topology, between gene sets related to different phenotypes. We demonstrated the utility of NetPAS in 50 well-curated gene sets and comparison of association studies using Z-scores, p-values or overrepresentations.Using NetPAS, a weighted human disease network was generated from the association scores of 19 gene sets from OMIM. We also applied NetPAS in gene sets derived from gene ontology and pathway annotations and showed that NetPAS uncovered functional terms missed by DAVID and other network-based enrichment tools. Overall, we show that NetPAS can take topological constraints of molecular networks into account and offer new perspectives than existing methods.

Protein interaction probability landscapes for yeast replicative aging

“…Similar method had been previously applied to a smaller yeast PIN and observed that the high-degree, hub proteins tend to avoid interactions with the other hub proteins, which was suggested to contribute to the stability of the PIN (Maslov and Sneppen 2002). This approach had been recently applied to estimate the association strengths between different categories of proteins using a human PIN (Guo and Qin 2020). Figure 5 shows the global landscape in the variable spaces of node degrees and RLS ratios of single-gene deletion mutations.…”

“…22,188 null network models were built in a way such that all vertices were randomly reshuffled, yet the node degrees preserved as the original PIN. We term these null models the MS02 null models (Maslov and Sneppen 2002;Qin et al 2003;Guo and Qin 2020).…”

“…For example, whereas based on the proteome the essential versus essential protein interactions fell into a deep basin in the PIPL of old age, based on the transcriptome these interactions were located on a high ridge. To understand the two opposite trends, another proteome-wide landscape was constructed based on comparisons between the empirical PIN and random, null network models (Guo and Qin 2020). In this landscape, we showed that unlike the hub proteins, the essential proteins favor interactions with other essential proteins.…”