Centrality of drug targets in protein networks

Ariele Viacava Follis

doi:10.1186/s12859-021-04342-x

Abstract

BackgroundIn the pharmaceutical industry, competing for few validated drug targets there is a drive to identify new ways of therapeutic intervention. Here, we attempted to define guidelines to evaluate a target’s ‘fitness’ based on its node characteristics within annotated protein functional networks to complement contingent therapeutic hypotheses.ResultsWe observed that targets of approved, selective small molecule drugs exhibit high node centrality within protein networks relative to a broader set of investigational targets spanning various development stages. Targets of approved drugs also exhibit higher centrality than other proteins within their respective functional class. These findings expand on previous reports of drug targets’ network centrality by suggesting some centrality metrics such as low topological coefficient as inherent characteristics of a ‘good’ target, relative to other exploratory targets and regardless of its functional class. These centrality metrics could thus be indicators of an individual protein’s ‘fitness’ as potential drug target. Correlations between protein nodes’ network centrality and number of associated publications underscored the possibility of knowledge bias as an inherent limitation to such predictions.ConclusionsDespite some entanglement with knowledge bias, like structure-oriented ‘druggability’ assessments of new protein targets, centrality metrics could assist early pharmaceutical discovery teams in evaluating potential targets with limited experimental proof of concept and help allocate resources for an effective drug discovery pipeline.

Highlights

In the pharmaceutical industry, competing for few validated drug targets there is a drive to identify new ways of therapeutic intervention
Proteins within each set were assigned to a broad ‘target class’ based on Gene Ontology (GO) [33] identifiers: channels and transporters, enzymes, G-protein coupled receptors (GPCRs), kinases, nuclear receptors (Fig. 1a, b)
30 40 50 60 70 degree degree as potential drug targets, with good agreement between source networks from the String0.7 and Proteome networks, respectively (2043 shared predicted targets for the entire network training, contingency probability < 1 0−4; 562 shared predicted targets for individual classes training, contingency probability < 1 0−4). We found that these four predictive models (String0.7 and Proteome, trained either over the entire database or in a class-specific fashion), correctly identified 406/503 (81%) of drug target proteins annotated in a recent study [3]

Summary

Introduction

In the pharmaceutical industry, competing for few validated drug targets there is a drive to identify new ways of therapeutic intervention. Pharmaceutical companies strive to select suitable targets and minimize attrition This has driven more than two decades long efforts towards the identification and annotation of ‘druggable’ fractions of the genome [1,2,3]. Many naturally existing networks, including biological signaling networks, exhibit an approximate [13] scale-free organization characterized by a power law dependence of their node degree distribution [14,15,16,17]. Scale-free organization results in short acrossnetwork distances and confers a network robustness to the perturbation of a limited number of its edges [18] These characteristics are intuitively advantageous to biological signaling as they help fulfill the conflicting requirements of efficient response to external stimuli (short distance) while preserving homeostasis upon perturbation (robustness) [15]. Each protein (node) has a specific function, hub proteins in signaling networks may play gateway roles at a higher hierarchical level [19]

Methods

Results

Conclusion