A Comparative Study of Cluster Detection Algorithms in Protein-Protein Interaction for Drug Target Discovery and Drug Repurposing.

Jun Ma,Penggao Dai,Benjamin Haibe-Kains,Jenny Wang,Laleh Soltan Ghoraie,Xin Men

doi:10.3389/fphar.2019.00109

Abstract

The interactions between drugs and their target proteins induce altered expression of genes involved in complex intracellular networks. The properties of these functional network modules are critical for the identification of drug targets, for drug repurposing, and for understanding the underlying mode of action of the drug. The topological modules generated by a computational approach are defined as functional clusters. However, the functions inferred for these topological modules extracted from a large-scale molecular interaction network, such as a protein–protein interaction (PPI) network, could differ depending on different cluster detection algorithms. Moreover, the dynamic gene expression profiles among tissues or cell types causes differential functional interaction patterns between the molecular components. Thus, the connections in the PPI network should be modified by the transcriptomic landscape of specific cell lines before producing topological clusters. Here, we systematically investigated the clusters of a cell-based PPI network by using four cluster detection algorithms. We subsequently compared the performance of these algorithms for target gene prediction, which integrates gene perturbation data with the cell-based PPI network using two drug target prioritization methods, shortest path and diffusion correlation. In addition, we validated the proportion of perturbed genes in clusters by finding candidate anti-breast cancer drugs and confirming our predictions using literature evidence and cases in the ClinicalTrials.gov. Our results indicate that the Walktrap (CW) clustering algorithm achieved the best performance overall in our comparative study.

Highlights

Drugs physically bind specific target proteins and activate downstream effectors to change the gene expression profiles of tumor cells, which are highly modular in the context of molecular interaction networks (Hartwell et al, 1999; Berger and Iyengar, 2009; Sah et al, 2014)
To investigate differences in clustering structure produced by the four clustering algorithms, CLE, CWlabel propagation (CLP), called the Walktrap (CW), and CI, we applied them to the MCF-7 cell-based protein–protein interaction (PPI) network, consisting of 7,904 proteins and 213,422 interactions
The number of small clusters was larger than the number of big clusters in all algorithms (Figure 2A), which comprise a smaller proportion of the total number of clusters (15%, CW; 27.6%, CLE; 29%, CLP and; 0.3%, CI) (Supplementary Table S3), consistent with a previous report that compared the modular structure of human cell-agnostic PPI networks generated by seven community detection methods (Liu et al, 2017)

Summary

Introduction

Drugs physically bind specific target proteins and activate downstream effectors to change the gene expression profiles of tumor cells, which are highly modular in the context of molecular interaction networks (Hartwell et al, 1999; Berger and Iyengar, 2009; Sah et al, 2014). Kenley and Cho proposed a graph entropy algorithm to identify functional clusters from PPI networks (Kenley and Cho, 2011). These efforts have led to more effective modeling of PPIs and the drug targeting thereof with respect to specific diseases

Methods

Results

Conclusion