Protein domain connectivity and essentiality

TL;DR

This study investigates the relationship between protein domain connectivity and essentiality, revealing that domains' connectivity correlates more strongly with essentiality than whole proteins, using statistical analysis on multiple network models.

Contribution

It introduces a domain-level analysis of protein interaction networks, showing a stronger correlation between connectivity and essentiality than protein-level analysis.

Findings

Domains have a higher connectivity-essentiality correlation than proteins.

Statistical analysis confirms the significance of domain-level connectivity.

Different network models support the robustness of the findings.

Abstract

Protein-protein interactions can be properly modeled as scale-free complex networks, while the lethality of proteins has been correlated with the node degrees, therefore defining a lethality-centrality rule. In this work we revisit this relevant problem by focusing attention not on proteins as a whole, but on their functional domains, which are ultimately responsible for their binding potential. Four networks are considered: the original protein-protein interaction network, its randomized version, and two domain networks assuming different lethality hypotheses. By using formal statistical analysis, we show that the correlation between connectivity and essentiality is higher for domains than for proteins.