Page 74 - GPD-1-2

Gene & Protein in Disease Recent advances and challenges of network biology

alternative, cost-effective choice in predicting the functions human viruses, and drugs [127] . There are also studies that
of genes/proteins. On the one hand, based on the guilt-by- have discussed the application of network control theory
association principle, the considered genes in a biological in identifying driver genes in cancer [133,134] . Interestingly,
network tend to have similar functions if the neighbored researchers have successfully predicted the functions of
genes have certain biological functions. On the other hand, neurons in C. elegans connectome based on structural
since biological networks have modularity structures, controllability [131] , thus providing a direct experimental
genes in the same module of a biological network tend to proof of the validity of these widely used control principles.
have similar functions [124,125] . In fact, many of the reviewed Network biology has important applications in network
works in Section 3.2.1. follow this rule, where the identified medicine [140-148] . On the one hand, network biology provides
important genes tend to have critical biological functions. effective tools for identifying drug targets [130,141-144] . In 2007,
For example, based on GDCN and the guilt-by-rewiring Yıldırım et al. built a bipartite graph composed of the
[29]
principle, our recent work predicted the functions of United States Food and Drug Administration-approved
several unannotated genes in Brassica napus [107] , including drugs and proteins linked by drug-target associations;
BnaC03g00220D and BnaC05g02200D. Other than that, the topological analyses of the constructed network
several researchers have attempted to predict the function revealed that many drugs target already targeted proteins.
of miRNA through the identification of miRNA targets; if By including drugs currently under investigation, they
the targets of a given miRNA are enriched in a biological identified a trend towards more functionally diverse targets
process or pathway, then it is reasonable to infer that the improving polypharmacology. The associated investigation
regulating miRNA is involved in that process .
[48]
is able to guide rational drug design. Besides that, from
Network biology enhances our understanding of many the perspective of structural controllability, Vinayagam
model organisms. For example, in Saccharomyces cerevisiae, et al. [130] reported that the structural controllability theory
researchers have detected pairwise genetic interactions can be used to identify disease genes and drug targets. The
among ~90% genes . It is reported that essential genes copy number alterations data of 1547 cancer patients were
[6]
of Saccharomyces cerevisiae are network hubs, displaying analyzed, and they found 56 indispensable genes, which are
5 times as many interactions as non-essential genes. The frequently amplified or deleted in nine different cancers.
set of genetic interactions or the genetic interaction profile However, 46 out of the 56 genes have not been previously
for a gene provides a quantitative measure of its function. reported to be associated with cancer. Recently, Valle et al. [142]
A global genetic interaction network underlines the developed a network medicine framework to show that the
functional organization of a cell and provides a resource proximity of polyphenol targets and disease proteins can
for predicting gene and pathway function. It is predicted predict the therapeutic effects of polyphenols. The work
that six poorly characterized genes – MTC2, MTC4, revealed the predictive power of network biology. On the
MTC6, CSF1, DLT1, and YPR153W – may function other hand, network biology has potential implications
as novel functional modules that are important for the in drug combinations and drug repurposing [145-148] . It has
growth of Saccharomyces cerevisiae in high-pressure and been widely reckoned that drug development for complex
cold environments . diseases is seeing a shift from targeting individual proteins
[6]
Network biology enables the clarification of molecular or genes to system-based attacks targeting dynamic
mechanisms behind certain phenotypes and the control of network states [140,143,144] . Based on the theoretical tools
biological systems [126-134] . On the one hand, following the developed in network biology, Cheng et al. [146] proposed a
identification of important genes and the prediction of network-based approach to identify clinically efficacious
their functions, KEGG pathway analysis or GO enrichment drug combinations for specific diseases. The proposed
analysis can be performed to analyze the functions of method facilitated the identification and verification of
gene clusters [130,135] . In particular, it is feasible to identify antihypertensive combinations, offering a powerful tool
the genes involved in crucial pathways that dominate the to identify efficacious combination therapies in drug
related biological phenotypes [107] . There are many tools for development [146] . Drug repurposing can effectively promote
enrichment analysis, including GSEA [135,136] , DAVID [137] , the processes in drug development. It is a cost-effective
ClusterProfiler [138] , GOEAST [139] , and so on. On the other approach to preventing human diseases, especially major
hand, based on the structural controllability theory of epidemic diseases. For drug repurposing, Cheng et al. [145]
complex networks, the identified functional genes can have identified hundreds of new drug-disease associations
be investigated from the perspective of control [126-134] . for over 900 FDA-approved drugs by quantifying the
Research has revealed that about 21% of proteins in human network proximity of disease genes and drug targets in
PPI network are indispensable. These indispensable the human PPI network. Most recently, several researchers
proteins are primary targets of disease-causing mutations, have looked into drug repurposing for COVID-19 [147,148] .

Volume 1 Issue 2 (2022) 8 https://doi.org/10.36922/gpd.v1i2.101