leukotriene A4 hydrolase (LTAH4-h), by designing a common pharmacophore

having the combined relevant features from both targets followed by structure-

based VLS (Wei et al. 2008).

2.3

Data Mining to Identify Novel Targets from “Big Data”:

A Network Systems Biology Approach

The explosion in the amount of biological data being generated and freely available

to the research community has shifted focus of attention to the development of new

techniques for data mining. Data mining involves retrieval, extraction and filtering of

valuable data from the “big data” available online, and in polypharmacology, target

identification was the first application of data. Ozgur et al. used support vector

machine (SVM) methods to construct a gene-disease interaction network and were

able to successfully confirm high association between the predicted candidate genes

and prostate cancer (Özgür et al. 2008). Similarly, other researchers have used data-

and structure-based data mining approaches to predict novel cancer targets and also

identify potential targets for cancer imaging and therapy (Pospisil et al. 2006, 2007).

Data mining has also been used to identify unknown relationships between genes

and disease using systems biology approaches to analyse polypharmacology. Cheng

and colleagues developed a web server, PolySearch, to provide related genes,

proteins, metabolites and drugs based on a given disease, or vice versa (Cheng

et al. 2008). Other data mining tools such as GeneWays that focuses on Alzheimer’s

disease and GenCLip are also based upon gene interactions present in molecular

networks (Krauthammer et al. 2004; Huang et al. 2008; Wang et al. 2014).

Polypharma, a novel database, has 953 ligands complexed with more than two

structures of distinct protein families in the RCSB Protein Data Bank (PDB). It

has provided some interesting insights into ligand-target interactions, such as multi-

target ligands are slightly more hydrophobic and tend to have lower molecular

weights (<200 Da) than single-target ligands (Reddy and Zhang 2013; Reddy

et al. 2014).

2.4

Drug Repurposing

A direct application of polypharmacology is drug repurposing/repositioning,

i.e. identifying a new clinical use for an existing approved drug (Ashburn and

Thor 2004; Aubé 2012). A closely related concept is drug rescue, as for the case

of sildenafil (Viagra) (DeBusk et al. 2004). In many instances, drug repurposing has

occurred by serendipity (Paolini et al. 2006), but now concerted efforts are being

made to conduct drug repurposing systematically by envisioning three general

strategies, namely, chemical, biological and data mining (Boran and Iyengar

2010). Drug repurposing is primarily a retrospective approach, which offers mani-

fold benefits to the pharmaceutical industry, such as lower drug development costs

and reduced time for approval, as shelved drugs can be quickly marketed for new

22

T. R. Sahrawat and R. C. Sobti