Random Network Behaviour of Protein Structures

TL;DR

This paper investigates the network of side-chain interactions in proteins, revealing that at a coarse level they resemble random graphs, which may explain functional diversity, while at a finer level they show unique structural features.

Contribution

It demonstrates that protein side-chain interaction networks largely follow random graph models, providing insights into their role in functional flexibility and structural specificity.

Findings

Side-chain networks resemble random graphs at a coarse level

Deviations from randomness reflect structural uniqueness

Network behavior relates to protein stability and function

Abstract

Geometric and structural constraints greatly restrict the selection of folds adapted by protein backbones, and yet, folded proteins show an astounding diversity in functionality. For structure to have any bearing on function, it is thus imperative that, apart from the protein backbone, other tunable degrees of freedom be accountable. Here, we focus on side-chain interactions, which non-covalently link amino acids in folded proteins to form a network structure. At a coarse-grained level, we show that the network conforms remarkably well to realizations of random graphs and displays associated percolation behavior. Thus, within the rigid framework of the protein backbone that restricts the structure space, the side-chain interactions exhibit an element of randomness, which account for the functional flexibility and diversity shown by proteins. However, at a finer level, the network exhibits deviations from these random graphs which, as we demonstrate for a few specific examples, reflect the intrinsic uniqueness in the structure and stability, and perhaps specificity in the functioning of biological proteins.